The candidate gene approach to understanding the genetics of human essential hypertension is discussed by analyzing the contribution of 2 genes, angiotensinogen (AGT) and epithelial amiloride-sensitive sodium channel (ENaC). From a large series of studies conducted in humans and animals, it appears that the AGT gene plays a significant but modest role in human blood pressure variance. Mutations of the beta- and gamma-ENaC subunits are responsible for Liddle's syndrome, but the implication of the 3 ENaC subunits in essential hypertension is still questionable. Several lessons can be learned from these studies and applied to other candidate genes in essential hypertension: (1) Many linkage or association studies have a limited statistical power; (2) The genetic findings may vary greatly according to the populations studied; (3) There is a need for better phenotyping of the hypertensive population; (4) The causal relationship between molecular variants and hypertension is and will be difficult to establish firmly; (5) The contribution of genetic studied in rodents to the molecular genetics of human hypertension must be re-examined; (6) Most molecular variants lead to a low attributable risk in the population or a low individual effect at the individual level; and (7) It is too early to propose dietary recommendations and specific drug treatment according to patients' genotypes.