The ventilatory response (V E ) to electrically induced rhythmic muscle contractions (ERCs) was studied in six urethane-chloralose-anaesthetized sheep, while arterial oxygen and carbon dioxide pressure (P a,O 2 and P a,CO 2 ) and perfusion pressure were maintained constant at the known chemoreception sites. With cephalic P a,CO 2 held constant, the response to inhaled CO 2 was virtually abolished (0.03 ± 0.04 l min −1 Torr −1 ). During low-current ERC, which doubled the metabolic rate (V CO 2 increased from 192 ± 23 to 317 ± 84 ml min −1 , P < 0.01),V E followed the change inV CO 2 closely (from 5.24 ± 1.81 to −9.27 ± 3.60 l min −1 , P < 0.01) in the absence of any chemical error signal occurring at carotid and central chemoreceptor level (∆cephalic P a,CO 2 = −0.75 ± 1 Torr). Systemic P a,CO 2 decreased by −2.47 ± 1.9 Torr (P < 0.01). Both heart rate and systemic blood pressure increased significantly by 18.6 ± 5.5 beats min −1 and 7.0 ± 9.3 mmHg, respectively. When the CO 2 flow to the central circulation was reduced during ERC by blocking venous return (V CO 2 decreased by 102 ± 45 l min −1 , P < 0.01), ventilation was stimulated (from 11.99 ± 4.11 to 13.01 ± 4.63 l min −1 , P < 0.05). The opposite effect was observed when the arterial supply was blocked. Finally, raising the CO 2 content and flow in the systemic blood did not significantly stimulate ventilation provided that the peripheral and central chemoreceptors were unaware of the changes in blood CO 2 /H + composition. Our results support the existence of a system capable of controlling blood P a,CO 2 homeostasis when the metabolism increases independently of peripheral and central respiratory chemoreceptors. Information from the skeletal muscles related to the local vascular response provides the central nervous system with a respiratory stimulus proportional to the rate at which gases are exchanged in the muscles, thereby coupling ventilation to the metabolic rate.