1. A resting sartorius muscle of the frog or toad possesses a special kind of elasticity which is shown to be due to a component lying between the two sets of filaments. The elastic effect is seen only for very small length changes, up to about 0.2% of the muscle length, and the ;elastic limit' is then reached. If the length change then continues at a constant velocity the tension developed is maintained at a fixed level, producing a sort of frictional resistance. The component responsible is called the ;short-range elastic component', or SREC.2. It is also shown that a small part of the permanent tension of a resting muscle is probably due to ;active' interaction between the filaments. This is called the ;filamentary resting tension', or FRT. For a sarcomere length of 2.0 mu the FRT amounts to about 150 mg in a muscle weighing 100 mg.3. The stiffness of the SREC and the magnitude of the FRT are shown to be related to one another. They are both increased, and may rise to high values, by making the external solution hypertonic.4. The working hypothesis is as follows. In a resting muscle the cross-bridges on the myosin filaments are not entirely inactive, but a very small proportion of them are cross-linked with the actin filaments. The links are very stable and have a long ;life'. The elastic behaviour is due to the flexural rigidity, or spring-like properties of these bridges. The elasticity is ;short-range' because the bridges can be bent, or stretched, only a small way from the steady-state position before the contacts ;slip'. The ;filamentary resting tension' (which is present in the absence of any external length change) is attributed to an ;active process', which operates by imparting ;organized' potential energy to the participating cross-bridges, by potentiating their attachment, against their elastic resistance, to sites on the actin filaments which are displaced towards the Z line.5. It is shown that the ability of the muscle to maintain a ;frictional' resistance to a continuing, slow, length change is suppressed during a maintained tetanic contraction. The process of suppression starts during the period of the latency relaxation.6. It is suggested that the latency relaxation may be due to a reduction of the filamentary resting tension.7. The experiments with the latency relaxation were facilitated by the use of strongly hypertonic solutions. The positive twitch tension is then greatly reduced, or may be practically abolished, while the latency relaxation remains at about its normal size and is extended in duration.