A study of the limit of active and passive knee extension in 64 healthy adults revealed a physiological quadriceps lag; that is, in most subjects the active limit of knee extension fell short of the passive limit. With the subjects seated, for the passive test the examiner lifted the heel until the relaxed knee sagged into full extension under its own weight. The active test component comprised maximum active extension held for at least 5 sec. Videotaped reference markers on the lateral aspect of the limb were computer-analysed to derive the active and passive test positions. The active limit of knee extension was less than the passive limit by an average 2.5 degrees at the instant of maximum active knee extension, and by 2.9, 3.5, 4.0, 4.5 and 5.0 degrees 1, 2, 3, 4 and 5 sec later. At 0 and 5 sec, 16% and 41% of the subjects manifested a quadriceps lag of at least 5 degrees. There was no correlation between the magnitudes of passive knee extension and quadriceps lag. Since clinicians typically take several seconds to estimate visually or otherwise measure knee extension, account should be taken of the duration of maximum active contraction, as well as other details of test methodology, if quadriceps lag tests are to produce valid and reliable results.