There is overwhelming evidence that genetic factors play a role in the predisposition to suffer the development of the chronic inflammatory bowel diseases. The genetic analysis of complex diseases, such as ulcerative colitis and Crohn's disease, is difficult. The presence of disease heterogeneity, the relative low frequency in the population, the degree to which first-degree relatives are affected (approximately 10%), the presence of genes with minor genetic effects, and ethnic differences are some of the difficulties encountered when identifying disease susceptibility loci. Two major approaches to identifying these genes are being followed at present. The first, family-based, consists of studying linkage analysis in sibling pairs and parental transmission in genome-wide screening using microsatellite markers. These studies are appropriate and helpful for finding genes of major or moderate effects but may provide difficulty when identifying genes with minor effects. Risch and Merikangas have pointed to the power of association studies utilizing candidate genes in families. These studies should be considered in the future in genome-wide screens when technologic advances permit. The second approach is based on classic epidemiologic designs, population-based studies, using candidate genes in the framework of a biologic hypothesis. Recent data using both approaches in both Crohn's disease and ulcerative colitis are reviewed. The results of genome-wide linkage studies have not reached consensus but suggest that these diseases are different and polygenic in nature. We have started our studies with the hypothesis that an abnormal immune dysbalance contributes to the biologic basis of disease. We therefore study polymorphisms in genes encoding proinflammatory and regulatory cytokines. Preliminary data of these association studies suggest the importance of several genes with small effects in determining the severity and prognosis of these diseases.