The physiological responses of 12 competitive swimmers to a standard training workout of four sets of six 100 m freestyle repetitions per set were measured in two trials. By random assignment the swimmers attempted to keep the workout constant in each trial by maintaining either a target heart rate (HR) or a target time for the 100 m efforts. The target HR or time was determined from these values measured at the anaerobic threshold (AT) during an initial determination of this exercise level. The HR after each repetition, 50 m split time, and blood lactate concentration after each set were measured throughout each workout for each swimmer. When target HR was used as the exercise criterion, the group mean (±SD) HR for the first set of six 100 m repetitions (173±11 beats/min) was significantly lower (p≤0.05) than the group mean AT target HR (182±11 beats/min), but the group mean HR for sets two, three, and four was not significantly different from the target value. Group mean lactate concentration after each of the four sets was not significantly different from the group mean value measured at the AT, although the group mean 100 m times for sets one and two were significantly faster (p≤0.05) than the speed corresponding to the AT. When target time was the exercise criterion, the group mean 100 m time in each of the four sets was virtually identical to the group mean target time, but the group mean HR and lactate concentration in each set were all significantly lower (p≤0.05) than the group mean values measured at the AT. It was concluded that target HR and target time were both reasonably, but not completely, successful as a method of controlling exercise intensity, because the predetermined physiological criterion response measured by the AT test were not replicated exactly during the workout because of the accumulating degree of fatigue experienced by the swimmers.