To determine the effects of chronic exercise on the coronary collateral circulation of dogs with normal coronary arteries, 1-yr-old purebred beagles were divided into sedentary control and exercising groups. The latter were trained to run on a treadmill. A lower maximal heart rate during a standardized exercise test protocol after a 10- to 12-week training period and a higher gastrocnemius cytochrome oxidase activity in the runners attested to the presence of cardiovascular and skeletal muscle training effects. However, left ventricular weights, left ventricle/body weight ratios, myocardial myofibrillar and myosin ATPases, and hemodynamics were similar in sedentary and exercising dogs except for a significantly higher resting cardiac output in the runners. After occlusion of the left anterior descending coronary artery, both collateral conductance (retrograde flow/aortic pressure) and collateral flow measured with microspheres tended to be lower in the trained dogs, but differences were not significant. The endocardial/epicardial flow ratio in the ischemic area after coronary occlusion did not distinguish between exercisers and controls. Thus treadmill running in the dog with normal coronary arteries produced a training effect, but had no effect on coronary collateral vessels.