The primary hyperoxalurias type 1 (PH1) and type 2 (PH2) are autosomal recessive calcium oxalate kidney stone diseases caused by deficiencies of the metabolic enzymes alanine:glyoxylate aminotransferase (AGT) and glyoxylate/hydroxypyruvate reductase (GR/HPR), respectively. Over 50 mutations have been identified in the AGXT gene (encoding AGT) in PH1, associated with a wide variety of effects on AGT, including loss of catalytic activity, aggregation, accelerated degradation, and peroxisome-to-mitochondrion mistargeting. Some of these mutations segregate and interact synergistically with a common polymorphism. Over a dozen mutations have been found in the GRHPR gene (encoding GR/HPR) in PH2, all associated with complete loss of glyoxylate reductase enzyme activity and immunoreactive protein. The crystal structure of human AGT, but not human GR/HPR, has been solved, allowing the effects of many of the mutations in PH1 to be rationalised in structural terms. Detailed analysis of the molecular aetiology of PH1 and PH2 has led to significant improvements in all aspects of their clinical management. Enzyme replacement therapy by liver transplantation can provide a metabolic cure for PH1, but it has yet to be tried for PH2. New treatments that aim to counter the effects of specific mutations on the properties of the enzymes could be feasible in the not-too-distant future.