An accepted measure of anaerobic capacity is the maximal O2 deficit. But it is not feasible to use O2 deficit if > or =10 submaximal runs are needed to extrapolate the O2 demand of high velocity running (Medbø et al. 1988). Recently, an alternative method to determine O2 deficit was proposed (Hill 1996) using only results of supramaximal cycle ergometer tests. The purpose of this study was to evaluate this alternative method with data from treadmill tests. Twenty-six runners ran at 95%, 100%, 105%, and 110% of their velocity at VO2max. Times to exhaustion, velocity, and accumulated oxygen uptake (VO2) from each individual's four tests were fit to the following equation using iterative nonlinear regression: accumulated VO2 = (O2 demand x velocity x time)-O2 deficit. The mean value s derived for O2 demand and O2 deficit were 0.198+/-0.031 ml x kg(-1) x m(-1) and 42+/-22 ml x kg(-1). SEE for the parameters were 0.007+/-0.007 ml x kg(-1) x m(-1) and 8+/-10 ml x kg(-1), respectively. Mean R2 was 0.998+/-0.003. It was concluded that O2 deficit can be determined from all-out treadmill tests without the need to perform submaximal tests.