The chromosome 6 markers, HLA-ABC, D, DR, MT, properdin factor Bf, and complement factors 2 (C2) and 5 (C4), were studied in three families, each of which included two HLA identical siblings, one or both of whom were known to be HLA-B: GLO recombinants. The families were also typed with primed lymphocyte typing (PLT) for HLA-D/DR region associated DP antigens. None of these studies gave evidence that the recombinations had occurred within the HLA region. Mixed leucocyte culture (MLC) tests within the families showed no detectable stimulation between the HLA identical siblings in two of the families, but a very weak stimulation between the HLA identical siblings (H and G) in the third family (GG). No reactive PLT reagents were generated when cells from the HLA identical siblings of the first two families were primed against each other. In contrast, priming between cells of H and G gave rise to reactive reagents. One of these (GHx), reacted with a determinant which segregated within the GG family as if child G was a paternal recombinant between the HLA-D, DR, DP, and C4 loci, on the one hand, and on the other hand one or more loci governing other HLA-D/DR region controlled lymphocyte activating determinants. This reagent was only restimulated by cells from two of 47 unrelated individuals. The other PLT reagent (HGx) did not give a clearcut pattern within the family because it was weakly positive with all family members (most of whom were D/DR2-positive) except the specific responder; in the panel it reacted with a determinant significantly associated with D/DR/DP2. Other PLT reagents could be generated within the family against lymphocyte activating determinants controlled by genes in the two paternal haplotypes telomeric to the assumed recombinational site. These reagents gave stronger reactions than the HGx and GHx reagents and reacted with two determinants in the unrelated panel strongly associated with D/DR/DP2 and D/DR/DP6, respectively. It seems likely that the GG family represents a third example of a recombination between the HLA-DR and SB loci. Our findings further support the assumption that the DR determinants may be immunodominant in lymphocyte activation.