The aims of this study were to assess changes in muscle architecture, isometric and dynamic strength of the leg extensor muscles, resulting from dynamic resistance training, and the relationships between strength and muscle architecture variables. The participants (n = 30) were randomly assigned to one of two groups. The training group (n = 16; age 21.8 +/- 2.3 years, body mass 74.8 +/- 9.2 kg, height 1.75 +/- 0.08 m) performed dynamic resistance training for 13 weeks. The control group (n = 14; age 19.9 +/- 1.5 years, body mass 74.0 +/- 8.5 kg, height 1.76 +/- 0.05 m) did not perform any resistance training. Maximal dynamic and isometric strength were tested in both groups, before and after the training period. The members of the training group used the free-weight squat lift (90 degrees ) as their training exercise. The concentric phase of the squat was performed explosively. Skeletal muscle architecture of the vastus lateralis was visualized using ultrasonography. At the end of the study, significant increases in vastus lateralis muscle thickness (+6.9%, P < 0.001), fascicle length (+10.3%, P < 0.05), one-repetition maximum (+8.2%, P < 0.05), rate of force development (+23.8%, P < 0.05) and average force produced in the first 500 ms (+11.7%, P < 0.05) were seen only in the training group. Adaptations to the muscle architecture in the training group limited the loss of fibre force, and improved the capacity for developing higher velocities of contraction. The architectural changes in the training group were similar to those seen in studies where high-speed training was performed. In conclusion, dynamic resistance training with light loads leads to increases in muscle thickness and fascicle length, which might be related to a more efficient transmission of fibre force to the tendon.