It has been suggested in a number of studies that susceptibility to adult Hodgkin's disease (HD) is influenced by the HLA class II region, and specifically by alleles at the HLA-DPB1 locus. Since HD is diagnostically complex, it is not clear whether different HLA-DPB1 alleles confer susceptibility to different HD subtypes. To clarify this we have extended a previous study to type DPB1 alleles in 147 adult HD patients from a single centre. We have analysed patients with nodular sclerosing (NS), mixed cellularity (MC) or lymphocyte predominant (LP) HD, and gender in relation to HLA-DPB1 type, in comparison with 183 adult controls. The results confirmed previously reported associations of DPB1*0301 with HD susceptibility (relative risk (RR) = 1.42; 95% confidence interval (CI) 0.86-2.36) and DPB1*0201 with resistance to HD (RR = 0.49; CI 0.27-0.90). However, analysis by HD subtype and gender showed that *0301-associated susceptibility was confined to females with HD (RR = 2.46; CI 1.02-5.92), and *0201-associated resistance to females with NS-HD (RR = 0.28; CI 0.10-0.79). Susceptibility to NS-HD was also associated in females with *1001 (RR = 11.73; CI 1.32-104.36), and resistance with *1101 (RR = 0.08; CI 0.01-0.65). In contrast, susceptibility to LP-HD was associated in males with *2001 (RR = 32.14; CI 3.17-326.17), and to MC-HD with *3401 (RR = 16.78; CI 2.84-99.17). Comparison of DPB1-encoded polymorphic amino-acid frequencies in patients and controls showed that susceptibility to MC-HD was associated with Leucine at position 35 of DPB1 (RR = 8.85; CI 3.04-25.77), Alanine-55 (RR = 15.17; CI 2.00-115.20) and Valine-84 (RR = 15.94; CI 3.55-71.49). In contrast, Glutamic acid 69 was significantly associated with resistance to MC-HD (RR = 0.14; CI 0.03-0.60). Certain DPB1 alleles and individual DPbeta1 polymorphic amino acid residues may thus affect susceptibility and resistance to specific HD subtypes. This may be through their influence on the binding of peptides derived from an HD-associated infectious agent, and the consequent effect on immune responses to the agent.