We previously reported that the tension response to a 20% ramp shortening consisted of an initial fast tension decline followed, typically, by a slow tension decline; the tension at the transition between the two phases (P₂) could be used to construct the force – shortening velocity curve (Roots & Ranatunga, 2005). The cause of the continued slow tension decline after P₂ transition, however, was not clear. Thin filament deactivation has been shown to occur during and after shortening but typically with sub-maximal activation (Edman, 1975 and refs therein). In skinned fibres, the increase of calcium level abolished this type of deactivation (Eklund & Edman, 1982) and, in intact fibres, such deactivation was reduced by caffeine (Edman, 1980). Millimolar caffeine is known to increase Ca²⁺ release and myofilament Ca²⁺ sensitivity; therefore, we have now examined the effects of caffeine on the tension response to ramp shortening in tetanized rat muscle fibres. Adult male rats (~250 g) were terminally anaethetised; small bundles (~5 fibres) were isolated from the flexor hallucis brevis (a fast muscle) of the foot and mounted horizontally between a force transducer and servo-motor at 20 ^oC. Initial fibre length (L₀, ~2 mm) was set for maximal tetanic tension; the sarcomere length was ~2.5 μm. Experiments were done when the fibre bundle was immersed in normal Ringer solution (control) and in Ringer containing 2.5–5 mM caffeine. A fibre bundle was tetanized and, on the tension plateau, a ramp shortening (up to 20% L₀ in amplitude and at velocities of 0.01 to ~5 L₀/s) was applied and the tension changes monitored. With 2.5–5.0 mM caffeine, the twitch tension was potentiated (>75%) indicating an increase in [Ca²⁺] release on activation; however, the tetanic tension was not significantly changed. In experiments on 5 fibre bundles, the mean (± s.e.m.) tetanic tension (in kN/m²) was 236 (± 24.4) in control solution before exposure to caffeine and it was 233 (± 27.1) after recovery; the tension in the presence of caffeine was 234 (± 25.7) and the differences were not significant (P>0.7, paired t test). The continued slow tension decline during ramp shortening was seen in the presence of caffeine; with increase of shortening velocity, the rate of tension decline increased to a similar extent both in the presence of caffeine and in the control condition. When the force–shortening velocity curves were constructed using the tension level at the P₂ transition and analysed using Hill equation, the V_max (in L₀/s) was 3.28 (± 0.32) before caffeine and 3.23 (± 0.44) after recovery from caffeine: the V_max with caffeine (3.33 ± 0.59) was not significantly different (P>0.6). There was also no significant difference in the a/P₀ ratio (Hill’s equation) indicating the curvature of the relation was similar between control and caffeine data. Thus, our data suggest that the tension response during ramp shortening is not significantly altered by caffeine; preliminary findings indicate that the tension decline is not seen at fibre lengths >15% longer than L₀.