The 20 Metre Multistage Shuttle Run Test (20 MST) was modified for application to ice-sports, more specifically for ice-hockey and figure-skating. Seventy two participants in ice-sports served as the total subject group. Subjects included in the study were National and Provincial standard male ice-hockey subjects (n=67) and female figure skaters (n=5) participating in the Gauteng area of South Africa (altitude of 1497 metres above sea level and barometric pressure of 655 mmHg). The mean age for the total group was 17.44±1.33 years. The research methodology entailed a repeated measures design to determine: a) velocity of motion on-ice vs. over-ground; b) energy expenditure on-ice vs. over-ground; and c) mechanical efficiency on-ice vs. over-ground. The mean velocity of motion measured over three distances (0 to 20, 0 to 30 and 0 to 40 m) indicated a significantly (p≤0.05) faster velocity on-ice (5.99±0.72 m/s) versus over-ground (5.75±0.63 m/s). The corresponding mean time-lapsed on-ice/over-ground ratio was 0.97±0.11. Differences in mean energy expenditure whilst performing the original 20 MST over-ground as opposed to on-ice were measured at low (at 4 minutes of exercise and 10 km/h), intermediate (after 8 minutes of exercise and 12 km/h), and high intensity (after 12 min of exercise and 14 km/h). The mean of the three indicated a significantly (p≤0.05) higher energy expenditure over-ground (14.04±4.86 kcal/min) as apposed to on-ice (10.51±2.95 kcal/min). The mean energy expenditure ratio for the three different intensities on-ice vs. over-ground was 0.74±0.21. Similarly, the mechanical efficiency index over-ground (4.92±0.59) was found to be significantly (p≤0.001) poorer than on-ice (6.83±1.49). The mean mechanical efficiency ratio over-ground/over-ice was 0.74±0.13. Subsequently, based on the above results, the 20 MST was modified by: a) adapting (increasing) the velocity of motion required for each level of the test (distance of 20 m per shuttle); and b) establishing the reliability and validity of the modified 20 MST for use on-ice. The adapted 20 Metre Multistage Shuttle Skating Test (the modified (skating) 20 MST) started at a velocity of 2.8 m/s (10.1 km/h) and permitted 7.1 seconds to complete each shuttle for the first level of the test, which then decreased progressively at each level. This was based on an over-all variable-derived on-ice to over ground ratio of 0.84. Test-retest, on-ice reliability measures (n=15) for predicted VO2max (49.5±8.37 vs. 49.29±7.95 ml/kg/min) showed a highly significant (p£0.001) consistency (r=0.87). Similarly test-retest concurrent validity measures (n=10) for predicted VO2max over-ground with the original 20 MST (48.09±6.25 ml/kg/min) as designed by Léger and Lambert (1982) versus on-ice values with the adapted on-ice 20 MST (49.98±7.23 ml/kg/min), showed a very significant (p£0.01) correlation of 0.73 between the two tests. In conclusion the original 20 MST, as designed by Léger and Lambert (1982) for over-ground, proved inappropriate for use on-ice. Modification of the starting velocity as well as a progressive increase in velocity for all subsequent stages renders the modified 20 MST for ice-sports a reliable and valid test for cardiorespiratory fitness (VO2max), with surface-specific utility. The 20 Metre Multistage Shuttle Run Test (20 MST) was modified for application to ice-sports, more specifically for ice-hockey and figure-skating. Seventy two participants in ice-sports served as the total subject group. Subjects included in the study were National and Provincial standard male ice-hockey subjects (n=67) and female figure skaters (n=5) participating in the Gauteng area of South Africa (altitude of 1497 metres above sea level and barometric pressure of 655 mmHg). The mean age for the total group was 17.44±1.33 years. The research methodology entailed a repeated measures design to determine: a) velocity of motion on-ice vs. over-ground; b) energy expenditure on-ice vs. over-ground; and c) mechanical efficiency on-ice vs. over-ground. The mean velocity of motion measured over three distances (0 to 20, 0 to 30 and 0 to 40 m) indicated a significantly (p≤0.05) faster velocity on-ice (5.99±0.72 m/s) versus over-ground (5.75±0.63 m/s). The corresponding mean time-lapsed on-ice/over-ground ratio was 0.97±0.11. Differences in mean energy expenditure whilst performing the original 20 MST over-ground as opposed to on-ice were measured at low (at 4 minutes of exercise and 10 km/h), intermediate (after 8 minutes of exercise and 12 km/h), and high intensity (after 12 min of exercise and 14 km/h). The mean of the three indicated a significantly (p≤0.05) higher energy expenditure over-ground (14.04±4.86 kcal/min) as apposed to on-ice (10.51±2.95 kcal/min). The mean energy expenditure ratio for the three different intensities on-ice vs. over-ground was 0.74±0.21. Similarly, the mechanical efficiency index over-ground (4.92±0.59) was found to be significantly (p≤0.001) poorer than on-ice (6.83±1.49). The mean mechanical efficiency ratio over-ground/over-ice was 0.74±0.13. Subsequently, based on the above results, the 20 MST was modified by: a) adapting (increasing) the velocity of motion required for each level of the test (distance of 20 m per shuttle); and b) establishing the reliability and validity of the modified 20 MST for use on-ice. The adapted 20 Metre Multistage Shuttle Skating Test (the modified (skating) 20 MST) started at a velocity of 2.8 m/s (10.1 km/h) and permitted 7.1 seconds to complete each shuttle for the first level of the test, which then decreased progressively at each level. This was based on an over-all variable-derived on-ice to over ground ratio of 0.84. Test-retest, on-ice reliability measures (n=15) for predicted VO2max (49.5±8.37 vs. 49.29±7.95 ml/kg/min) showed a highly significant (p£0.001) consistency (r=0.87). Similarly test-retest concurrent validity measures (n=10) for predicted VO2max over-ground with the original 20 MST (48.09±6.25 ml/kg/min) as designed by Léger and Lambert (1982) versus on-ice values with the adapted on-ice 20 MST (49.98±7.23 ml/kg/min), showed a very significant (p£0.01) correlation of 0.73 between the two tests. In conclusion the original 20 MST, as designed by Léger and Lambert (1982) for over-ground, proved inappropriate for use on-ice. Modification of the starting velocity as well as a progressive increase in velocity for all subsequent stages renders the modified 20 MST for ice-sports a reliable and valid test for cardiorespiratory fitness (VO2max), with surface-specific utility. / Dissertation (MA (Human Movement Science))--University of Pretoria, 2005. / Biokinetics, Sport and Leisure Sciences / unrestricted