The purpose of this study was to determine the intrasession and intersession reliability of EMG vastus medialis oblique:vastus lateralis (VMO:VL) ratios at four knee positions (0, 45, 60, and 90°) at 100 and 60% of maximal voluntary isometric contraction (MVIC). Once reliability was established, the second purpose was to determine VMO:VL ratios and torque at each knee position. Thirty-two subjects participated in two sessions; 19 subjects were tested at 100% MVIC and 13 were tested at .60% MVIC. Results revealed the following intraclass correlations: 100% MVIC intrasession .40-.80, intersession .40-.70; 60% MVIC intrasession .60-.90, intersession .50-.80. A significant difference in torque occurred at all knee positions except 60 versus 90°. No significant difference existed in VMO:VL ratios at the four positions of knee flexion. Pain and measurement error significantly increased during 100% MVIC testing. It was concluded that no selective VMO recruitment occurred as revealed by the VMO: VL ratios in asymptomatic subjects. Future study is needed that reports EMG reliability data during exercises that theorize selective VMO recruitment.