One of the most important assumptions of the passive method of gravity correction in isokinetic dynamometry is that subjects are relaxed during the test. The purpose of this study was to examine the validity of the passive gravity correction method in pubertal children. Fifteen pubertal males performed passive knee extensions and flexions at 30°·s
$^{-1}$
. The electromyographic (EMG) activity of the vastus medialis (VM), vastus lateralis (VL) and biceps femoris (BF) was recorded using bipolar surface electrodes. The GM was estimated using five different methods: the recorded GM during knee extension and flexion, the estimated GMs from the recorded knee extension and flexion GMs at 30° of knee flexion and the GM estimated using anthropometric data. The results indicated that the passive knee movements resulted in significantly (p<0.05) higher GM compared to those estimated using anthropometric data or mathematically using the GM at 30°. The GM in the range of 31--90° was significantly (p<0.05) higher during knee extension compared with knee flexion. The VM and VL raw EMGs were significantly (p<0.05) higher during knee extension compared to knee flexion movement, reaching 40% of the their maximum EMG values. The BF EMG ranged from 7% to 10% of the maximum, but it was not significantly different between extension and flexion phases. It can be suggested that the recorded dynamometer moment during passive gravity correction tests is affected by forces exerted by musculotendinous structures which are activated during the test. The application of passive gravity correction techniques in pubertal children demonstrates significant problems and should be applied with extreme caution.
