We determined whether the increment in cardiac end-diastolic compliance (a reduced diastolic stiffness constant) following endurance training is related to alterations in myocardial collagen characteristics. Sixteen weeks of habitual exercise (Ex) in rats, which produced left ventricular (LV) hypertrophy (LVH) [LV weight in g: Ex = 1.01 (0.04), sedentary control = 0.89 (0.04); P < 0.05], resulted in a reduced LV end-diastolic (LVED) chamber stiffness [slope of the linearised LVED pressure versus LVED internal diameter relation in kPa x mm(-1): Ex = 0.67 (0.03), control = 0.80 (0.03); P < 0.05]. The increased LVED chamber distensibility was associated with an attenuated myocardial stiffness [slope of the linearised LVED stress versus strain relation in g x cm(-2); Ex= 15 (3), control = 25 (2); P < 0.05]. Although LV total collagen content (mg) was increased in the exercised rats [Ex = 5.0 (0.3), control = 4.1 (0.2); P < 0.05], this was a reflection of the presence of LVH, as the myocardial collagen concentration (microg x mg(-1) LV wet weight) was unaltered [Ex = 4.9 (0.2), control = 4.6 (0.2)]. Furthermore, habitual exercise did not influence the percentage of myocardial collagen extracted following cyanogen bromide digestion (an index of collagen cross-linking), [i.e. Ex = 38 (3), control = 38 (3)], nor the proportion of myocardial collagen phenotypes I and III [I/III; Ex = 3.04 (0.20), control = 2.85 (0.22)]. In conclusion, exercise-induced increments in end-diastolic myocardial distensibility are unlikely to be a consequence of alterations in the properties of myocardial collagen.