Fatigue rates of the vastus medialis oblique and vastus lateralis muscles were compared from static and dynamic exercise. Based upon clinical observations, it was expected that the vastus medialis oblique would demonstrate greater fatigability than the vastus lateralis. However, based upon physiological and morphological considerations, it was expected that the vastus lateralis would exhibit greater fatigability than the vastus medialis oblique. In the static exercise condition, nine subjects--having no history of knee problems--maintained an isometric knee extension torque at 30 and 60% of their maximum value until exhaustion, during which time electromyography (EMG) data were collected every 10 s. The knee angle was 20 degrees flexion. In the dynamic exercise condition, a different group of seven subjects performed exercise sets consisting of eight cycles of concentric-eccentric knee extension, with a resistance equal to 40% of the maximum isometric value. The range of motion was restricted to the terminal 30 degrees of knee extension. Each set of eight repetitions was followed by a 1 s 50% maximum isometric knee extension, during which time EMG data were collected. The raw EMG data were numerically processed to extract the median frequency of the power density spectrum, which has been shown to reflect the metabolic processes associated with fatigue. Linear regression generated a slope coefficient representing the rate of change of the median frequency, with respect to contraction duration for each subject, muscle, and condition. Analyses of variance, with repeated measures from both exercise conditions suggest that "short-arc" quadriceps exercise did not selectively fatigue either the vastus medialis oblique or vastus lateralis, thereby supporting neither clinical nor physiological expected outcomes. Therefore, the clinical contention that short-arc quadriceps exercises selectively strengthen the vastus medialis oblique is questioned.