The relation between sarcomere length, tension and time course of tension development in twitch and tetanic contractions at 20 degrees C was determined for isolated fibres from the semitendinosus muscle of the frog (Rana esculenta). In twenty fibres at about 2.15 micron sarcomere length, the peak twitch tension, the maximum tetanic tension and the twitch/tetanus ratio ranged, respectively, from 0.22 to 1.6 kg/cm2, from 2.13 o 3.96 kg/cm2 an from 0.07 to 0.53. The peak twitch tension was found to be: i) directly correlated with the twitch/tetanus ratio and the time to the peak of the first derivative of the twitch tension, ii) inversely correlated with the time to the peak of the first derivative of tetanic tension. No significant correlation was found between the maximal tetanic tension and the peak twitch tension or the twitch/tetanus ratio. Peak twitch tension and twitch/tetanus ratio were not correlated with the fibre cross-sectional area which ranged from 1.052 to 6,283 micron2. Sarcomere length-tension curves for twitch and tetanic isometric contractions at 20 degrees C were determined in twelve fibres. Increases in sarcomere length from about 2.15 to 2.85 micron produced, depending on the peak twitch tension or the twitch/tetanus ratio at about 2.15 micron, either decrease and no change or increase in peak twitch tension, but constantly enhanced the twitch/tetanus ratio and the degree of this potentiation was inversely correlated with the twitch/tetanus ratio at 2.15 micron. Increase in sarcomere length above 2.15 micron did not alter the course of the early development of twitch and tetanic tensions, reduced considerably the variation in peak twitch tension and twitch/tetanus ratio, without altering that of tetanic tension and swamped the correlation between the peak twitch tension and the time to peak of the differentiated twitch tension. However, the peak twitch tension at about 2.85 micron resulted to be directly correlated with the peak twitch tension at about 2.15 micron and in addition the relative length-dependent change in the time of the peak of the first derivative of the twitch tension resulted to be directly correlated with the relative length-dependent change in the peak twitch tension. It is concluded that both the duration of the active state and the rate factors of activation contribute to the determining of the large variation in peak twitch tension at about 2.15 micron, whereas the length-dependent increase in twitch/tetanus ratio appears to be mainly determined by prolongation of the active state duration.