Prior constant-load exercise performed for 30-min at or above maximal lactate steady state (MLSSp) significantly impairs subsequent time-to-task failure (TTF) compared with TTF performed without prior exercise. We tested the hypothesis that TTF would decrease in relation to the intensity and the duration of prior exercise compared with a baseline TIP trial. Eleven individuals (6 males, 5 females, aged 28 +/- 8 yrs) completed the following tests on a cycle ergometer (randomly assigned after MLSS p was determined): (i) a ramp-incremental test; (ii) a baseline TIT trial performed at 80% of peak power (TTFb); (iii) five 30-min constant-PO rides at 5% below lactate threshold (LT-5%), halfway between LT and MLSSp (Delta(50)), 5% below MLSSp (MLSS-5%), MLSSp, and 5% above MLSSp (MLSS+5%); and (iv) 15- and 45-min rides at MLSSp (MLSS15 and MLSS45, respectively). Each condition was immediately followed by a TIP trial at 80% of peak power. Compared with ITFb (330 +/- 52 s), there was 8.0 +/- 24.1, 23.6 +/- 20.2, 41.0 +/- 14.8, 52.2 +/- 18.9, and 75.4 +/- 7.4% reduction in TIP following LT-5%, Delta(50), MLSS-5%, MLSSp, and MLSS+5%, respectively. Following MLSS is and MLSS 45 there were 29.0 +/- 20.1 and 69.4 +/- 19.6% reductions in TIT, respectively (P < 0.05). It is concluded that TIT is reduced following prior exercise of varying duration at MLSS p and at submaximal intensities below MLSS. Novelty: Prior constant-PO exercise, performed at intensities below MLSSp, reduces subsequent TTF performance. Subsequent TTF performance is reduced in a linear fashion following an increase in the duration of constant-PO exercise at MLSSp.