Recently, fatigability and muscle oxygen consumption (mVO(2)) during sustained isometric contractions were found to be less at shorter (30 degrees knee angle; 0 degrees = full extension) compared to longer knee extensor muscle lengths (90 degrees ) and, at low torques, less in the rectus femoris (RF) muscle than in the vastus lateralis and medialis. In the present study we hypothesized that these findings could be accounted for by a knee angle- and a muscle-dependent activation respectively. On two experimental days rectified surface EMG (rsEMG) was obtained as a measure of muscle activation in nine healthy young males. In addition, on day 1 maximal torque capacity (MTC) was carefully determined using superimposed nerve stimulation on brief high intensity contractions (> 70%MVC) at 30, 60 and 90 degrees knee angles. On day 2, subjects performed longer lasting isometric contractions (10-70%MTC) while mVO(2) was measured using near-infrared spectroscopy (NIRS). At 30 degrees , maximal mVO(2) was reached significantly later (11.0 s +/- 6.5 s) and was 57.9 +/- 8.3% less (average +/- SD, across intensities and muscles) than mVO(2) at 60 and 90 degrees (p < 0.05). However, rsEMG was on average only 18.0 +/- 11.8% (p = 0.062) less at the start of the contraction at 30 degrees . At 10%MTC at all knee angles, maximal mVO(2) of the RF occurred significantly later (28.8 +/- 36.0 s) and showed a significantly smaller increase in rsEMG compared to both vasti. In conclusion, it is unlikely that the tendency for less intense muscle activation could fully account for the approximately 60% lower oxygen consumption at 30 degrees , but the later increase in RFmVO(2) seemed to be caused by a less strong activation of the RF.