The three mutations that have been linked to osteopetrosis cause defects in the acidification of bone. The most common of these, found in 50 to 60 percent of patients, results in defects in the A3 subunit of the osteoclast vacuolar H+-ATPase proton pump. The second most clinically significant mutation affects CLCN7, a gene encoding an osteoclast-specific chloride channel. These mutations occur in 10 to 15 percent of patients with severe autosomal recessive osteopetrosis and have been implicated in "intermediate" and autosomal dominant osteopetrosis as well. Carbonic anhydrase II dysfunction is a feature of autosomal recessive osteopetrosis but accounts for a small proportion of patients with osteopetrosis. Several patients have been reported with the equivalent of the murine grey-lethal mutation, but this mutation also occurs in few children with osteopetrosis. It should be noted that a substantial percentage of patients with osteopetrosis have no identifiable gene defect. Before these molecular abnormalities were found, clinical descriptions were the exclusive means of characterizing osteopetrosis. Classifications based on molecular events and associated physiology will undoubtedly be more precise. It is logical to expect that there is a role for hematopoietic stem-cell transplantation in patients with intrinsic osteoclast defects and severe osteopetrosis. Further investigations will facilitate correlation of the various genotypes with the clinical presentation, anticipated complications, prognosis, and expected response to treatment. This information will provide a basis for making more informed decisions regarding the care of patients with osteopetrosis.