Rats were injected intraperitoneally with 1 mCi (each) of [3H]lysine at Day 11 of neonatal life to label their lung collagen. Five weeks later, half of the animals were given an intratracheal injection of 1.5 U of bleomycin sulfate via a tracheostomy; control animals received saline intratracheally by the same technique. Age-matched groups of control and bleomycin-treated rats were killed, and their lung collagen was analyzed at zero (control animals only), 1, 2, 4, 6, and 10 wk after bleomycin administration, a time course appropriate for development of pulmonary fibrosis in this animal model. We measured radioactivity in hydroxylysine and in the difunctional collagen crosslinks hydroxylysinonorleucine and dihydroxylysinonorleucine at each time point. No evidence of breakdown of this pool of mature, preformed collagen was observed in lungs of either the control or the bleomycin-treated rats. We also measured the total lung content of hydroxypyridinium, a trifunctional collagen crosslink, by its intrinsic fluorescence. There was no evidence of collagen degradation in lungs of either group of rats by this criterion either. We conclude that there is no biochemically detectable turnover of mature lung collagen, defined as that pool of lung collagen that is obligatorily extracellular (i.e., crosslinked and containing labeled hydroxylysine from an injection of precursor 5 to 15 wk earlier), in either normal rat lungs or lungs of rats made fibrotic with bleomycin. Statistical analysis of the data suggests that our methodology was sensitive and precise enough to have detected turnover of less than 0.5% of lung collagen per day, some 20-fold less than estimates of lung collagen turnover that have been suggested to be occurring in vivo by others using different techniques and presumably studying different pools of lung collagen.