Studies of the influence of dietary selection pressures in living and extinct primate taxa frequently interpret cranial diversity using a simple lever model. When this model is applied to functional or evolutionary questions, it is commonly assumed that the muscles of mastication vary little in activity during biting at points along the tooth row. A pattern of smoothly increasing maximum bite force magnitudes is therefore predicted as the bite point is moved posteriorly along the dental arcade. Diverse adaptive explanations are mapped onto this pattern. In this study, the activity of the superficial masseter and anterior temporalis muscles in humans was quantified during high magnitude bite force production at points along the tooth row. These data indicate that there are substantial changes in muscle activity with bite point, and that the standard lever model is therefore an incomplete description of masticatory force production. Maximum muscle force magnitudes were found to be greatest during first molar biting and to decrease as the bite point moved anteriorly and posteriorly. Additionally, relative balancing and working side muscle activity changed by bite point. This latter pattern is consistent with the predictions of Greaves' constrained lever model, which assumes that masticatory muscle activity is restricted by the need to maintain compressive forces at both temporomandibular joints. However, these results also imply that additional factors influence muscle activity--such as dental morphology, mandibular kinematics, and the need to safeguard against joint distraction during diverse loading conditions--and that the constrained lever model of Greaves is therefore also incomplete. These considerations suggest that masticatory system morphology in primates will respond differently to dietary selection pressures than is commonly hypothesized. Intepretations of cranial morphology in fossil taxa may therefore also differ.