Determination of magnetic resonance (MR) relaxation times (T1 and T2) and ultrasound backscatter are potential methods of noninvasive myocardial tissue characterization. The authors postulated that infarct-related collagen deposition in a canine chronic myocardial infarction model would alter T1 and T2 relaxation times. Further, we sought to compare the changes in T1 and T2 with backscatter measurements. Eight animals were studied 4 to 6 weeks after occlusion of the left anterior descending coronary artery; standard echocardiography showed each to have left ventricular wall motion abnormalities. Time-averaged backscatter and cyclic backscatter variation (at 5 MHz) were measured in the open-chest animal with corresponding normal and infarcted areas undergoing in vitro T1 and T2 measurements (at 20 MHz) and water and hydroxyproline concentration analysis. Significant increases in T2, time-averaged ultrasound backscatter, and hydroxyproline concentration were found in the infarcted myocardium compared with normal muscle. Cyclic variation in backscatter and water content were not significantly different between the two groups. Although backscatter, T2, and tissue collagen content increased in the infarcted myocardium, a significant relationship was not found among these variables.