ABSTRACTLIEBER, D. C., R. L. LIEBER, and W. C. ADAMS. Effects of run-training and swim-training at similar absolute intensities on treadmill VO2max.Med. Sci. Sports Exerc., Vol. 21, No. 6, pp. 655–661, 1989. Thirty-seven sedentary males, aged 28–35 yr, were either run-trained, swim-trained, or served as controls in an 11 1/2-wk training study. Runners and swimmers exercised once a d, 3 d-wk, at a heart rate (HR) intensity equivalent to 75% of their treadmill VO2max. Treadmill maximal oxygen consumption (VO2max), submaximal cardiorespiratory response, and body composition parameters were measured before and following the training period. Runners, swimmers, and controls experienced a significant increase in treadmill VO2maxover the 11 1/2-wk study period. The 28 and 25% increases observed for the runners and swimmers, respectively, were significantly greater than the 5% increase observed for the controls (P< 0.0001). Runners and swimmers did not differ significantly from each other with respect to this increase in VO2max; nor did they demonstrate significant changes in respiratory exchange ratio (RER) at VO2maxbetween tests. The run-trained and swim-trained groups both experienced a decrease in HR at a standard submaximal walking workload but did not differ significantly from each other. Controls showed no significant change in submaximal exercise response. A significant difference was observed among groups (P< 0.01) for change in percent body fat. Changes in lean and fat weight over the training period were significant for both the runners (P< 0.002) and swimmers (P< 0.03) but not for the controls. Taken together, these data do not refute the concept of training specificity but do place restrictions on the conditions under which training specificity may be demonstrated. For example, under the training conditions in this study, central cardiovascular stresses were nearly equivalent, and therefore experimental groups, although trained in different exercise modes, demonstrated improvements in treadmill VO2maxwhich were significant, equivalent, and not mode specific.