Using an open circuit system (Douglas bag method), measurement of the anaerobic threshold (AT) was performed on ten healthy male college students during an incremental exercise test on a bicycle ergometer in an attempt to determine the validity of this method as compared with arterial blood lactate AT measurement. Blood samples were taken from either the radial or brachial artery through a Teflon catheter (3 ml/each time) every minute until the subject's maximal exercise tolerance was reached. Blood lactate was analyzed by the enzymatic method. Differences in work rate, VO2, %VO2max, VE, HR, and R at ATLA(AT determined by the increase in blood lactate) and at ATGE (gas exchange AT based on the non-linear increases in VE, VCO2, and other respiratory parameters), respectively, were all found to be statistically insignificant. There was a significant correlation (r = 0.866, p less than 0.01) between ATLA and ATGE when expressed i VO2 values (l/min). There was also a significant correlation between ATLA and VO2max (r = 0.778, p less than 0.01). These results indicate that the commonly used Douglas bag method could provide a valid non-invasive measure of anaerobic threshold.