While haemoconcentration due to loss of plasma volume is well established during cycling, the existence of similar changes during running remains contentious. This study compared the changes in plasma volume and associated blood indices during 60 min of running and cycling at the same relative intensity (approximately 65% VO2max), with all changes referenced to blood indices obtained after 30 min seated at rest on a cycle ergometer. Plasma osmolarity increased similarly with both forms of exercise but was less than predicted for water loss alone, such that there was a net loss of sodium during exercise and of potassium postexercise, with essentially no loss of protein. Plasma volume decreased similarly (approximately 6.5%) in both exercise trials, but while that with cycling was initiated by exercise itself and was essentially maximal within 5 min, the reduction in plasma volume in the running trial was induced by adopting the upright posture and was complete before exercise began. These data would indicate that different mechanisms are responsible for the changes in plasma volume induced by running and cycling, while the similarity of change would suggest that there is a lower limit to any reduction in plasma volume, regardless of mechanism. Furthermore, the observation that the changes in plasma volume were complete before or early in exercise, would imply that oral water ingestion during prolonged exercise, which is essential for thermoregulation, may be more concerned with homeostasis of extravascular water rather than plasma volume.