The heart rate of crustaceans changes with variations in ambient temperature within the normal environmental range (Maynard, 1960). The temperature coefficient (Q10) of the heart rate of crabs over the range 4­19 °C is about 2 (Florey and Kriebel, 1974). There are few studies of the heart response to a rapid change in temperature, although aquatic crustaceans often meet with warm or cold water masses (Spaargaren and Achituv, 1977). Electromechanical coupling of muscle fibres becomes less effective with decreasing temperature (Dudel and Ruedel, 1968), but a mechanism has been described that compensates for the tonus effect during leg muscle activity (Fischer and Florey, 1981). Compensatory mechanisms may also exist for heart muscle, and I have recently found that myocardial cells of a marine lobster begin to produce large action potentials in response to cooling. Lobster myocardial fibres develop tension in response to excitatory junction potentials (EJPs) generated by impulse activity of motor neurones in the cardiac ganglion (Van der Kloot, 1970; Anderson and Cooke, 1971; Kuramoto and Kuwasawa, 1980; Kuramoto and Ebara, 1984a). The heart tension produced is fed back to the cardiac ganglion because the cardiac neurones are sensitive to filling pressure (Maynard, 1960; Kuramoto and Ebara, 1984a, 1885, 1988, 1991). Thus, the responses of the isolated heart to cooling will result from the combined activities of the cardiac ganglion and the muscle cells. This report focuses on the development of a spiking response by the myocardial cells when the heart is cooled. The spikes produced correspond to enhanced contractions of the myocardium, suggesting that the myocardial cells may use this as a mechanism to compensate for the reduced efficacy of excitation­contraction coupling that occurs with falling temperature. Lobsters (Panulirus japonicus Von Siebolt, both sexes, approximately 200 g, N=25) were reared in an indoor aquarium continuously supplied with fresh natural sea water. Seasonal changes of aquarium temperature ranged from 15 to 25 °C. The isolated hearts were subjected to cooling experiments. The rate of cooling ranged from 1 to 3 °C min-1, the magnitude from 1 to 6 °C and the duration from 5 to 6 min. The methods for perfusing and recording from the isolated hearts were substantially the same as those used previously (Kuramoto and Ebara, 1984a, 1985, 1988, 1991). The perfusion saline was switched to warm or cold. Bath temperature near the heart was monitored with a platinum sensor (1 k omega at 0 °C). Myocardial membrane potentials were measured with glass microelectrodes (3 mol l-1 KCl, 10­30 M omega). Muscle tension was recorded using a strain gauge.