The purpose of this paper is 1) to describe maximal anaerobic alactacid (AAC) and lactacid (ALC) capacity test and 2) to determine their reliabilities in men and women. The subjects were submitted to either a 10-s (2 trials) or a 90-s (1 trial) all-out ergocycle test for AAC and ALC respectively. Thirty-four male and 24 female subjects were tested for AAC, while 21 males and 19 females took part in the ALC test. A modified bicycle ergometer allowed the exact measurement of the distance and the work load for the computation of the work performed. load for the computation of the work performed. Each subject was tested and retested within 7 days. In both AAC and ALC, male subjects performed more work than women. AAC was 108 +/- 16 (mean +/- SD) and 90 +/- 14 J/kg for males and females respectively while ALC was 486 +/- 50 and 377 +/- 34 J/kg. Although the work load was designed to be 0.09 kp/kg for the AAC and 0.05 kp/kg for the ALC tests, there were wide variations between subjects with respect to the optimal load (AAC: from 0.05 to 0.11; ALC: from 0.03 to 0.06 kp/kg). Reproducibility was consistently high, with intraclass correlations of 0.98 and 0.99 for AAC (AAC-Max) and ALC respectively, with no difference between the male and female subgroups. It is concluded that these AAC and ALC tests, designed under assumptions of face validity, allow for differences between males and females and are highly repeatable.