The aim of this study was to measure serial changes in the rate of blood lactate clearance (gamma2) in response to sequential periods of training and detraining in four male triathletes aged 22-44 years. There were two major phases of training and taper, each lasting 4-5 weeks (training 1 = 5 weeks, taper 1 = 2 weeks, training 2 = 4 weeks and taper 2 = 2 weeks), in preparation for a triathlon competition. The training stimulus absorbed by each subject was carefully quantified from the duration and intensity of the training exercise. A serial weekly measure of each trainee's physical response to training was evaluated as the peak power, termed a 'criterion performance', developed by a subject during a 30 W x min(-1) ramp cycle ergometer test to exhaustion each week. During 30 min of recovery after this test, 13 samples of venous blood were drawn sequentially from a subject to measure the blood lactate recovery curve. The rate constant of blood lactate clearance was estimated by a non-linear least-squares regression technique. In addition, the concurrent time to peak lactate concentration and the peak lactate concentration were also estimated to help define changing lactate kinetics. The criterion performance generally declined throughout each period of incremental training and improved during each taper period, rising iteratively in this way to be clearly above baseline by the end of the second taper. The blood lactate clearance rate increased transiently in early training before declining from the middle of the first training period to the middle of the first taper; thereafter, gamma2 increased above baseline in each trainee throughout the remaining first taper and the major portion of the second training period, decreasing only in the final criterion performance test. The time to peak lactate declined from baseline throughout all phases of training and taper. Peak blood lactate increased in all subjects to the end of the first taper before declining by the end of the second training period, rising again to baseline levels during the second taper. The change in gamma2 was examined relative to the work rate achieved in cycle ergometry above an initial baseline score (deltaCP) and against concurrent peak blood lactate. There was a clear upward shift in gamma2 above baseline throughout the first and second training and taper in two subjects; this was less clear in the remaining two subjects, each of whom had a lower deltaCP. We conclude that this indicates improved lactate clearance, manifest by the change in gamma2 induced by endurance training.