The effects of chronic exposure to ozone on lung collagen crosslinking were investigated in two groups of juvenile cynomolgus monkeys exposed to 0.61 ppm of ozone 8 hrs per day for 1 year. One group was killed immediately after the exposure period; the second exposed group breathed filtered air for 6 months after the ozone exposure before being killed. Previous studies of these monkeys had revealed that lung collagen content was increased in both exposed groups (J.A. Last et al., (1984). Toxicol. Appl. Pharmacol. 72, 111-118). In the present study specific collagen crosslinks were quantified in order to determine whether the excess collagen in the lungs of these animals was structurally normal or abnormal. In the group killed immediately after exposure, the difunctional crosslink dehydrodihydroxylysinonorleucine (DHLNL) was elevated, as was the ratio of DHLNL to dehydrohydroxylysinonorleucine (HLNL). Lung content of the mature nonreducible crosslink hydroxypyridinium was also increased in this group. In the group killed after a 6-month postexposure period, lung content of the difunctional crosslinks DHLNL and HLNL was indistinguishable from control values. However, lung hydroxypyridinium content was significantly increased. The changes in collagen crosslinking observed in the group killed at the termination of exposure are characteristic of those seen in lung tissue in the acute stage of experimental pulmonary fibrosis. The changes seen in the postexposure group suggest that while the lung collagen being synthesized at the time the animals were killed was apparently normal, "abnormal" collagen synthesized during the period of ozone exposure was irreversibly deposited in the lungs. This study suggests that long-term exposure to relatively low levels of ozone may cause irreversible changes in lung collagen structure.