A wide variety of techniques have been developed to monitor the mechanical responses of isolated cardiac myocytes. The most successful are those that measure shortening in unattached cells. Because of their relative ease of implementation, edge-detector methods of following cell displacement have become most widespread. Laser diffraction techniques have been applied to the single heart cells, and sophisticated sarcomere imaging systems capable of 2-ms time resolution of shortening responses have also been developed. Active force has been recorded in intact single cells from frog atria; however, the compliance of the force transducers was relatively high (~5% L(o)). (There is an obvious trade-off between transducer sensitivity, which affects noise and drift and compliance.) Some success has been reported with the use of intact rat myocytes supported by suction micropipettes and in guinea pig ventricular myocytes adhering to poly-L-lysine-coated glass beams. With the rat preparation, contractile stress was comparable to that of ventricular muscle, but few cells survived the attachment. In guinea pig myocytes, contractile stress in electrically induced twitches was only ~10% of the active stress developed by mammalian trabeculae or papillary muscles at the same temperature (35°C), but, as with the frog atrial transducer, the compliance of the supporting beams was relatively high. Sarcomere uniformity has not been evaluated in these intact preparations. For attachment to the relatively short mammalian cardiac myocytes, the more promising methods that better preserve sarcomere uniformity include double-barreled micropipettes coated with a barnacle adhesive; however, for nonsubmersible transducers, a continuing limitation is the problem of solution surface stability. Unfortunately, the more severe limitation to effective attachment to intact cells is still the extreme sensitivity of the sarcolemma to mechanical stress. The challenge remains to develop an attachment to the intercalated disk such that cell stress can be transferred to the supporting transducers along the normal stress-bearing cellular interface. The ultrastructural and passive mechanical data strongly indicate that although the extracellular collagen limits the extension of cardiac muscle beyond the peak of the active length-tension relation, there is also a substantial cellular component of resistance to extension. Furthermore, this cellular component is related to the cytoskeleton rather than to membranous elements in the cell. The more likely candidates for the longitudinal resting stress-bearing element are titin (connectin) and desmin. Titin is the most likely candidate, since it is present in both skeletal and cardiac muscle in relative large quantities, and immunolabeling and stiffness data show that titin extends from the M line to the Z band. These data also show that there is a close association of titin with the thick filaments in both cardiac and skeletal muscle and that passive stiffness falls as myosin is extracted and the titin attachment to myosin is disrupted in high salt. It remains to be established whether the influence of passive stress applied to the thick filaments by titin has any regulatory effect on active force development. It is apparent that the basis of the ascending limb of the Frank-Starling length-tension relation may be a composite of several factors, each of which may influence the magnitude of shortening in unattached cells when contractility is changed. The sensitivity of activating Ca2+ bound to TnC on cross-bridge attachment may play a predominant role in the length-tension relation. The development of restoring forces with cell shortening appears to be significant, but their effect may be more in restoring diastolic length than in influencing the length-tension relation. The relative contributions of other factors (cytosolic viscosity, cross-bridge compression, diastolic Ca2+) remain to be evaluated.