Quinolone antimicrobial agents induce two types of histological changes in the articular cartilage of synovial joints in immature, but not mature, laboratory animals. The first is cavity formation in the middle zone of the articular cartilage in animals of all species examined so far, and this can be induced by a single or several daily doses. The second is osteochondrotic lesions in the caudal femoral condyles of rats when subchronic or chronic treatment is started at a juvenile age, but there seems to be only one reference for this effect. The development process of cavity formation is as follows. Degeneration and necrosis of chondrocytes are first observed, and the surrounding matrix becomes edematous with demasked collagen fibrils and decreased safranin O stainability. A flat cleft or cavity is then formed in the edematous cartilage, and erosion is sometimes produced by detachment of the cavity outer wall. The potency of quinolones to form chelate complexes with Mg 2+ is discussed as causal for cavity formation. This could lead to Mg 2+ deficiency in the cartilage, resulting in integrin alteration and further radical formation that can finally induce cartilage lesions. On the other hand, an early change of osteochondrosis is a thickened middle zone of the articular cartilage with a thinned deep zone. The thickened cartilage protrudes into the epiphysis and prevents age-dependent thinning of the cartilage. In advanced cases, fissures are formed in the bottom of the thickened cartilage, and extension of the fissures to the cartilage surface induces detachment of the cartilage, subchondral bone necrosis and fibrotic lesions in the marrow space. The lesions are considered to be caused by quinolone-induced inhibition of the differentiation of chondrocytes in the middle zone into those in the deep zone. For chondrotoxicity in juvenile animals, the pediatric use of quinolones is generally contraindicated, but whether or not chondrotoxicity could really occur in pediatric patients is still controversial. (J Toxicol Pathol 2008; 21: 123-131)