Experiments were performed to better understand the relationship between glutaraldehyde and calcification of bioprosthetic heart valves, using both the cusps and the wall of porcine aortic roots. The results of the first experiment, for which 3H-labeled glutaraldehyde solutions were used, indicated that binding of glutaraldehyde in cusps and wall is concentration-dependent, that the wall contains significantly less glutaraldehyde than the cusp, and that glutaraldehyde, which penetrates in the wall at similar rates from the intima and the adventitia, is homogeneously distributed throughout the wall after 7 days of fixation, except for the intima side, where it is significantly lower. The results of the second experiment, for which cusps and 1-cm2 pieces of wall from glutaraldehyde-fixed porcine aortic roots were implanted subdermally in young rats, indicated that for both types of tissue, calcification appears to first initiate predominantly in the cell nuclei before extending to the other structures. After 8 weeks of implantation, whereas the cusps were completely calcified, calcification of the wall was limited to two longitudinal bands 150-300 microns thick, located below the adventitia and intima surfaces. The results of the third experiment indicated that cusp calcification, which decreased significantly after a 12-month storage period, was reset to high levels by reexposing the valves to glutaraldehyde at the end of the 12-month storage period. Wall calcification remained constant under all tested conditions. The results suggest that the mechanism(s) of calcification in the wall and the cusp may be different, and that calcification may be related to a particular molecular configuration resulting from exposure to glutaraldehyde.