Mechanisms responsible for the differences in humoral immune response to GAT (a random linear amino acid polymer) were investigated in a line of chickens consisting of four sublines homozygous for Ea-B (B1 or B19) and high or low antibody response to GAT (Ir-GATH or Ir-GATL). Previous research provided evidence of chromosomal recombination between the serologically determined regions of the MHC (encoded by B-F and B-G genes) and the gene or genes that control immune response to GAT, but immune response to GAT did not seem to be mediated through differences in B-L gene products. In the present study, proliferation of GAT-primed T lymphocytes indicated that reactivity in vitro was not associated with antibody levels produced in the animal. Cell surface markers were identified by flow cytometry. Lymphocytes from Ea-B19 chickens that were Ir-GATL had a higher percentage of suppressor T (CD8)-positive cells than did lymphocytes from Ir-GATH chickens. The Ea-B1 chickens that were Ir-GATL had a higher percentage of CD4-positive lymphocytes than did chickens that were Ir-GATH. This may indicate that low response to GAT in the Ea-B19 chickens, but not in Ea-B1 chickens, is mediated by CD8-positive cells. The ability of antigen-presenting cells (APC) to process and present GAT to antigen-primed T lymphocytes was tested in vitro. Measurements of lymphocyte proliferation indicated that, within the Ea-B1 blood type, APC from Ir-GATL chickens produced higher (P < .05) stimulation of both GAT low- and GAT high-responder lymphocytes. It is possible that, between the two B blood types, there are different mechanisms responsible for the differential response to GAT.