1. Single fibres isolated from the anterior tibialis muscle of Rana temporaria were allowed to shorten against a high load during a 2.5-4.0 s fused tetanus (1-3 degrees C) and the maximum force produced at the short length was compared with that recorded during a fixed-end tetanus at the same overall fibre length. Changes in length of marked, consecutive segments (ca 0.5 mm in length) along the fibre were measured throughout the tetanus using a photoelectric recording system. 2. Loaded shortening (load ca 3/4 of maximum tetanic force) starting from approximately 2.55 microns sarcomere length and ending near slack fibre length depressed the tetanic force by 13 +/- 2% (mean +/- S.E.M., n = 10) and caused a marked redistribution of sarcomere length along the fibre. Unloaded shortening over the same range caused no force deficit and did not lead to increased dispersion of sarcomere length. 3. Loaded shortening below slack length produced less force depression and less non-uniformity of sarcomere length than did a corresponding intervention above slack length. 4. The force deficit after loaded shortening, both above and below slack fibre length, was positively correlated (P < 0.005) to the coefficient of variation of the sarcomere length along the fibre. 5. The decrease in active force after loaded shortening, and its relation to increased dispersion of sarcomere length along the fibre, could be simulated closely by a computer model in which the muscle fibre was assumed to consist of eleven discrete segments acting in series with a passive elastic element. 6. Experiments were performed in which the length of an individual segment of the intact muscle fibre was strictly controlled throughout a tetanus. Loaded shortening of such a 'length-clamped' segment caused no force depression during the subsequent isometric phase either above or below slack fibre length. 7. The results suggest strongly that force depression after loaded shortening of a single muscle fibre is attributable to non-uniform sarcomere behaviour along the fibre. The experimental evidence supports the view that: (i) the myosin cross-bridges act as independent force generators; and (ii) their steady-state performance during a tetanus is unaffected by the preceding contractile activity.