This review of advances made toward understanding the molecular basis of congenital heart disease covers studies on subjects ranging from atrioventricular septal defects to zebrafish models. Genetically abnormal mice with atrioventricular septal defects have abnormal endocardial cushion development with the delayed appearance of mesenchymal cells and certain critical adhesion proteins. The prevalence of 22q11 deletions among patients with the conotruncal defects was estimated at 8% to 17%. Deletions were rare among patients lacking typical DiGeorge syndrome (DGS) or velocardiofacial (VCF) dysmorphic features, and more common in tetralogy of Fallot with pulmonary atresia than tetralogy of Fallot alone. Studies with patients with unusual 22q11 defects revealed that regional effects on several genes seem to underlie these complex phenotypes. A second DGS/VCF region on chromosome 10p13 was defined molecularly. Laterality defects (heterotaxy) have been associated with connexin43 mutations, and mice lacking connexin43 developed pulmonary atresia with intact ventricular septum. Three other groups failed to find connexin43 mutations in heterotaxy patients, suggesting genetic heterogeneity. Studies of cardiac looping with lower vertebrates revealed the critical role of the notochord. Ellis-van Creveld syndrome, an autosomal dominant skeletal dysplasia with atrial septal defects, and familial total anomalous pulmonary venous return, an autosomal dominant trait with reduced penetrance, were genetically linked to chromosomal bands 4p16 and 4p13-q12, respectively. The zebrafish has emerged as an important model for the study of the earliest stages of the cardiovascular system, and the miles apart and gridlock mutants, which have failure of heart tube fusion and aortic atresia, respectively, are discussed.