A number of studies have demonstrated aggregation of cases of ulcerative colitis or Crohn's disease in families, and of cases of both diseases within the same families, suggesting that patients share a genetic background. Perhaps because of differences in the selection of patients, study design and diagnostic criteria, different patterns of occurrence of inflammatory bowel disease (IBD) have been found among relatives of patients with these disorders. In recent years, however, several studies have been carried out, aiming by epidemiological methods to reveal (1) the frequency of familial occurrence of IBD among patients with ulcerative colitis and Crohn's disease, and (2) the prevalence of IBD among 1(0) relatives to patients with these diseases. Results from these studies show a relatively uniform pattern of family occurrence in about 10% of patients with ulcerative colitis and Crohn's disease, and a prevalence among 1(0) relatives of about 10 times that of the background population. A twin study reported a significantly higher concordance rate for Crohn's disease than for ulcerative colitis in monozygotic twins. By use of complex segregation analyses in 3 different studies, a very similar model of inheritance was found to fit for ulcerative colitis, namely a major dominant or additive gene with a low penetrance. For Crohn's disease the best-fitting model was a major recessive gene, with a high penetrance. This difference strongly supports the concept of ulcerative colitis and Crohn's disease as two separate disease entities. The occurrence of both diseases within the same families in certain members of the affected families is difficult to explain. The search for distinct associations of HLA genes with inflammatory bowel disease has shown a positive correlation between DR2 and ulcerative colitis and a negative association with DR4 and DRw6, compared with ethnically matched controls. In contrast, in Crohn's disease a positive association with the combination of DR1 and DQw5 alleles was revealed, thus indicating genetically different disease susceptibility for the two disorders. In general, however, no consistent pattern has been revealed from studies of association of HLA-A or -B antigens or blood group and serum protein markers. In two French families with several members affected with Crohn's disease no evidence for an HLA haplotype association could be revealed. Possible inherited markers of ulcerative colitis or Crohn's disease have been sought but without convincing success. Increased intestinal permeability, presence of anticolon antibodies and presence of antineutrophil leukocyte antibodies have been proposed, but not proved. Thorough studies are now needed of multimember families with disease for linkage studies to identify loci which contribute to increased liability. Such studies are in progress in different centres.