Thermoelastic behaviour similar to that of rubber has been found in numerous biological systems: elastin fibres, heat-contracted collagenous and elastoidin fibres, resting muscle and fresh 'silkworm'. In all these cases the elastic force must be attributed to the thermal motion of segments of partially stretched flexible chain molecules. We believe therefore that, generally, whenever high reversible extensibility is observed in living matter this latter contains a network of irregularly coiled or folded chain molecules. Such extensibility has been observed, for example, in animal membranes and in the elastic parts of cytoplasm. Several elastic systems contract on raising the temperature even if they are not subjected to a stress. These must contain two elastic systems which oppose each other: a system which contracts if the temperature is raised (such as does rubber under stress) and another one which is not affected in the same way by variation of temperature and maintains the stress of the first system. Such systems may be obtained by introducing cross-links in a stretched elastomer; in the case of collagenous fibres this can be achieved by treating with formaldehyde. Probably native elastin fibres contain a similar double system; they are not soluble in any of the solvents for proteins but become soluble under the action of elastase which seems to split up the cross-links. The 'long-range' elasticity of hair (and wool) keratin has given rise to a controversy; according to Astbury, Speakman and Woods, the elastic force is due to the tendency of the long '$\beta $' keratin modification to transform into the '$\alpha $' keratin modification which, according to these authors, has a lower energy. Other workers believe that the force is due to thermal motion. Now hair and wool are not highly extensible when dry; the measured stress is due to short-range cohesive forces. It is only after these attracting forces between neighbouring groups are sufficiently weakened by swelling agents that keratin becomes extensible. Then, after relaxation, the 'long-range' force, due to thermal motion, eventually becomes predominant. The $\alpha $-$\beta $ transformation is thus not the cause of the 'long-range' retracting force, which has the same origin as that observed with other elastomers. The contracting force of tetanized muscle cannot be attributed to a 'rubber-like' system. The stress-strain curve of tetanized muscle indicates that attracting forces are involved in muscular contraction.