A cylindrical segment, free of complex atherosclerotic lesions, was resected at autopsy from each of 59 descending human thoracic aortas by cutting just below the level of the first pair of intercostal arteries and 35 mm distal to this incision. Each isolated tunica media was defatted and subjected to successive treatment with EDTA-Tris, 5 M guanidine hydrochloride-Tris, 5 M guanidine hydrochloride-Tris-DTE, collagenase and either trypsin or hot alkali. After each extraction or digestion, the dimensions and weight of the segments were measured and the extracted materials were analyzed and quantitated. This allowed the total content of the various components of the tunica media to be assessed by both gravimetric and analytical means. An age-related rise was observed in the total content of the following components: proteins and glycoproteins soluble in chaotropic solvents (ranging from 24 mg/cm in the youngest samples to 46 mg/cm in the oldest) and collagen (38 mg/cm to 69 mg/cm). In contrast, the total content of elastin remained constant at 70 mg/cm at all ages, but its concentration decreased due to the rise in the concentration of the other tissue components as the tunica media thickened with age. It was also noted that with increasing age there was an accumulation of protein(s) which could not be solubilized by extraction with chaotropic agents or with collagenase, but which could be removed by treatment with either trypsin or hot alkali. Mechanical measurements conducted before and after trypsin digestion on samples previously subjected to purification with the first four agents used suggest that this accumulated protein(s) influenced the elastic response of the tissue to the applied stress by increasing the incremental modulus, the breaking stress, and the hysteresis. After the removal of this additional protein(s), the mechanical behavior of the elastin component was found to be identical in all samples, irrespective of age. It is therefore proposed that the morphological changes and the stiffening observed in the aging aortic wall are not due to degradation of its elastin network but to variations in the supramolecular organization of connective tissue components.