The development of inhibitors to factor VIII in patients with haemophilia A remains as a serious complication of replacement therapy. An apparently analogous condition has been described in a canine model of haemophilia A (Giles et al., Blood 1984; 63: 451). These animals and their relatives have now been followed for 10 years. The observation that the propensity for inhibitor development was not related to the ancestral factor VIII gene has been confirmed by the demonstration of vertical transmission through three generations of the segment of the family related to a normal (non-carrier) female that was introduced for breeding purposes. Haemophilic animals unrelated to this animal have not developed functionally significant factor VIII inhibitors despite intensive factor VIII replacement. Two animals have shown occasional laboratory evidence of factor VIII inhibition but this has not been translated into clinical significant inhibition in vivo as assessed by clinical response and F.VIII recovery and survival characteristics. Substantial heterogeneity of inhibitor expression both in vitro and in vivo has been observed between animals and in individual animals over time. Spontaneous loss of inhibitors has been observed without any therapies designed to induce tolerance, etc., being instituted. There is also phenotypic evidence of polyclonality of the immune response with variable expression over time in a given animal. These observations may have relevance to the human condition both in determining the pathogenetic factors involved in this condition and in highlighting the heterogeneity of its expression which suggests the need for caution in the interpretation of the outcome of interventions designed to modulate inhibitor activity.