Osteopetroses are rare human genetic disorders due to markedly decreased bone resorption. To date, the only gene whose inactivation was known to be responsible for human osteopetroses was that encoding carbonic anhydrase type II. Recessive malignant osteopetrosis is linked to decreased osteoclast function, unlike several osteopetroses in rodents that are caused by the inactivation of genes stimulating osteoclast differentiation. Recent advances in genetics have shown that some patients affected with recessive malignant osteopetrosis have inactivating mutations in a subunit of the vacuolar proton pump that is actively produced in the osteoclast brush border, but not in the lysosomes of other cells. The same gene is mutated in osteopetrotic oc/oc mice. The genes responsible for autosomal dominant osteopetrosis (ADO) have not yet been identified. Also, different localizations have been observed for ADO II, the type with sandwich vertebrae, and ADO I, presenting with diffuse osteocondensation. Less data than in malignant osteopetrosis are avaible on the cellular mechanism of decreased bone resorption in ADO but there is also genetic heterogeneity in ADO II. Pycnodysostosis is also due to a decreased resorption, and is quite close to osteopetrosis. Pycnodysostosis is linked to an inactivating mutation in the gene encoding cathepsin K, which is required for osteoclastic resorption. Decreased bone resorption is not the only defect inducing osteosclerosis, and Camurati-Engelmann disease is due to increased bone formation. Recently it has been shown that it is associated with a mutation in the propeptide of TGFbeta1. Thus, human osteosclerosing disorders have a wide range of phenotypes and genotypes and knowledge of them will contribute to a better understanding of the remodeling of normal bone.