Contact theory holds that increased contact between members of different groups can improve intergroup attitudes and reduce prejudice if it meets certain quality criteria, such as equal status, common goals, and cooperation within the contact situation. It is now emerging that even extended contact, or the mere knowledge that an ingroup member has a close relationship with an outgroup member, can lead to more positive intergroup attitudes. However, in South African society optimal contact is uncommon because most social spaces and structures are governed by strong norms of microsegregation that ensure that genuine optimal contact is uncommon. Given that multiracial families exemplify most features of optimal contact while radically challenging social norms of segregation, they may be important nodes for effective prejudice-reducing contact in South African society. This exploratory study investigated the extent to which general contact with people of other races, direct contact with multiracial families (i.e. personally knowing a member of a multiracial family), and extended contact with multiracial families (i.e. knowing someone who knows a member of a multiracial family) predicted reduced prejudice, reconciliatory race-policy attitudes and reduced intergroup threat. The results show that all three forms of contact predicted significantly reduced prejudice and that direct contact with multiracial families significantly predicted reduced prejudice over and above the effect of general interracial contact. Therefore we argue that contact researchers should consider more carefully whether normatively supported contact or radically norm-violating forms of contact optimally disrupt practices of microsegregation. Additionally, we argue that multiracial romantic relationships and families deserve increased attention and support as important nodes of prejudice reduction in South African society. However, general contact was also associated with reduced support for race-based social policies which supports the argument that the prejudice-reducing effects of interracial contact may be at least partially offset by other effects that may hinder broader social change.