Although the control of breathing during exercise has been studied extensively that both neural and humoral factors may play a role in mediating the exercise hyperpnea, it remains unresolved whether the response is due solely to neural or humoral mechanisms and how the pulmonary gas exchange matches to the tissue exchange rates. At just onset of exercise ventilation can increase abruptly within 1 to 2 breathes (Phase I). Such a rapid increment of ventilation is not attributed to the humoral factor(s) due to the circulation delay from muscle metabolites. More rapid mechanism(s) such as irradiation from brain stream, afferent neurogenic control, blood flow increment as cardiodynamic theory might be considered. During moderate exercise intensity with a step function, ventilation increased exponentially to the steady state (phase II) after an abrupt increase at just onset of exercise. In this phase II, the breathing is controlled as to exchange the oxygen and carbon dioxide at rates appropriate for the acid-base homeostasis regulation. As a result, arterial blood gases and acid-base status must maintain within a narrow range despite of extremely enlarged metabolic rates by exercise. Some humoral factor(s) might be mediated for the control of breathing such as mean steady state error theory, arterial blood gases oscillation theory, feedback system by CO2, or optimization theory. At the heavy exercise intensity, anaerobic metabolism was supplemented to product for the enhanced energy requirement, so that ventilation increased non-linearly with oxygen uptake due to the lactoacidosis and its buffering activity. There has been a debate concerning cause and effect relationship between blood lactate increment and hyperpnea during exercise. Recently, arterial potassium concentration resulted from depolarization due to muscle contraction during exercise might stimulate to carotid bodies for increasing ventilation.