Mechanical responses to stretch and length-tension relations were examined in rabbit taenia coli, mesenteric vein, aorta, and myometrium and in guinea pig taenia coli made atonic by incubation in Krebs-bicarbonate solution at 20-22 degrees C. When stretched 10% of the length at which maximum active tension is observed (Lo) in 0.5 s, the muscles showed a transient large force (resistance to stretch) that decayed to a new constant level within minutes (stress relaxation). The resistance to stretch decreased markedly in Ca2+-free [disodium ethylene glycolbis-(beta-aminoethylether)-N,N-tetraacetic acid (EGTA)] Krebs but was restored in normal Krebs solution. Calcium removal did not affect the passive length-tension curve. The absence of Ca2+ did not change the steady-state force maintained by the muscle; thus stretch resistance was not due to tone. Blockade of Ca2+ influx associated with electrical activity with 5-[3,4-dimethoxyphenethyl)methylamino]-2-(3,4,5-trimethoxyphenyl-2-isopropylvaleronitrile (D-600) and of Ca2+ release from intracellular sites with thymol (1 mM) completely blocked contraction but did not alter the responses to stretch, thus dissociating the responses to stretch from these processes and tension development. The Ca2+-dependent stress relaxation showed a dependence on muscle length similar to that for active tension development. Except at long muscle lengths, where connective tissue markedly affects length-tension relations, most of the "viscoelasticity" of these smooth muscles is dependent on calcium and may be largely due to the straining of crossbridges that are attached, but not generating a net force, in the resting state.