Metabolic costs and the pattern of upper extremity muscular fatigue for arm lifts were examined to test the hypothesis that for jobs requiring upper extremity lifts, the vertical and horizontal location of the end point of the lift are as important as the object weight. The tests were conducted with five healthy males, 22 to 34 years of age, inexperienced at industrial lifting jobs. The repetitive task, at five lifts per minute, was to grasp a weighted cylinder in each hand, assemble them into one unit, and place the unit on a rack. A 1 hour test series was performed for several height and reach distances and cylinder weights. Upper body fatigue was measured by electromyography (EMG) and tests of strength decrement. Whole body fatigue was assessed using heart rate, oxygen uptake, and measures of perceived fatigue. Muscle fatigue was assessed by changes in amplitude and frequency of the EMG signal. Heart rate, oxygen consumption, and static strength were measured before and after the lifting sessions. The experimental design included eight test conditions; one condition was run per day and at least 2 days elapsed between test conditions. Muscle fatigue results showed 32.2 and 14.7 percent average increases in the EMG amplitude for the 80 and 40 percent maximum voluntary contraction (MVC) weight conditions, respectively. The percentage decrease in static strength as a function of task conditions was 9.5 and 5.9 percent for the 80 and 40 percent MVC weight conditions, respectively. The effect of height was most significant on the biceps. The critical task condition was reach. The authors conclude that there are physiological bases for avoiding extreme reach and height, especially for weights in the range of 80 percent MVC. Because weight is a significant variable, an acceptable weight should be less than 80 percent MVC for all heights and reaches. Tasks should avoid combining excessive weight and reach requirements, heavy loads and extreme heights, or weights above 40 percent MVC along with excessive reach and height.