Studies were undertaken to determine the cardiac and ventilatory responses of restrained and unrestrained rainbow trout (Salmo gairdneri) to changes in inspired oxygen and carbon dioxide tensions. The role of blood oxygen carrying capacity in the control of ventilation was investigated, as well as the location and innervation of oxygen receptors activated by hypoxia. Ventilation volume (Vg) was measured directly in restrained fish using a ventilation chamber that separated inspired from expired water, the latter being collected in a graduated cylinder. In receptor localization experiments a wooden tongue depressor held vertically in the buccal cavity in the median plane divided water flows to the gills on each side of the fish. Thus, one set of gills could be irrigated with hyperoxic water to maintain arterial oxygen tension, while hypoxic water was passed over the other set of gills. Blood samples were obtained from cannulae implanted in both the dorsal aorta and right common cardinal vein. Vg increased in hypercapnea (inspired CO₂ tension (PICO₂) 0.5-2.0 kPa) due to increased "stroke volume (frequency remained constant), with higher levels of Vg recorded at higher C0₂ tensions. In fish exposed to PICO₂ levels of 0.5 and 0.9 kPa, raising the inspired oxygen tension (PIO₂) to 60.4 kPa eliminated the ventilatory response to hypercapnea. Hyperoxia had little or no effect on ventilatory responses to (PICO₂) levels of 1.5 and 2.0 kPa. Ventilation volume was inversely related to blood oxygen content (CaO₂) in trout. CaO₂ decreased and Vg increased during hypercapnea (PICO₂ 0.8 kPa), hypoxia (PIO₂12.4 kPa) and anaemia (haematocrit reduced from 22.3% to 14.3%), while CaO₂ increased and Vg decreased during hyperoxic hypercapnea (PIO₂ 60.4 kPa, PICO₂ 0.8 kPa). Increased Vg during hypercapnea is attributed to hypoxaemia produced by Bohr and Root off-shifts which result from increased blood CO₂ tension and reduced blood pH. Oxygen uptake remained constant during all experimental trials, indicating that the manoeuvre of increasing Vg is effective in relieving adverse effects of hypoxaemia. The significance of elevated Vg as a short-term adaptation to hypoxaemia is discussed, Heart rate decreased and ventilation increased in unrestrained fish exposed to gradual hypoxia (PIO₂ decreased from 20 kPa to 4 kPa) at 7°C and 16°C. The initial heart rate of fish acclimated to 16°C was higher than that of the 7°C group, but at the lowest level of PIO₂, heart rates of both groups dropped to the same level. Thus, the cardiac chronotropic response to hypoxia in trout is temperature independent. Receptors causing hypoxic bradycardia are located in the dorsal region of the first gill arch. Hypoxic bradycardia was eliminated by removing the first gill arch, or by sectioning the branches of cranial nerves IX and X innervating the arch. Blood flow through the arch does not appear to be necessary for this response, since ligation of the arch at its ventral insertion on the body wall did not affect hypoxic bradycardia. The pseudobranch has no role in cardiac control since interrupting the flow of blood through, and deafferentation of, the pseudobranch had no effect on the cardiac response to hypoxia. The biological significance of hypoxic bradycardia, and ventilatory-circulatory interaction during hypoxia, are discussed. Ventilatory responses to hypercapnea and hypoxia were unchanged after bilateral section of the nerves to the first gill arch. Receptors in the first gill arch thus have no role in control of ventilation during either hypercapnea or hypoxia. Possible locations for receptors responsible for control of ventilation are discussed. / Science, Faculty of / Zoology, Department of / Graduate