1. Anodonta cygnea L. in captivity exhibits rhythmical behaviour as follows:
(a) Periods of activity and quiescence alternate, with frequencies varying with different specimens from 3 to 30 per week. During the periods of quiescence the shell valves are completely adducted. During active periods the valves normally gape. (The ‘slow’ rhythm.)
(b) During the active periods the adductor muscles show frequent rapid contractions followed by slow relaxations. This activity is also rhythmical, and has a frequency of up to 20 per hour. (The ‘rapid’ rhythm.)
2. The neuromuscular mechanism of these rhythms has been investigated, and it is shown that: (a) both adductor muscles participate in the rhythms; (b) the slow rhythm is a function of the unstriated portions of the muscles; (c) the rapid rhythm is a function of the striated portions of the muscles; (d) the rapid rhythm in each adductor muscle is controlled by its nearest ganglia; (e) the slow rhythm in each adductor muscle is controlled by the combined effect of the nearest ganglia, which tend always to produce a tonus, and the cerebropleural ganglia, which at intervals inhibit that tonus; (f) the control of the rhythms appears to be intrinsic in the nerve ganglia, and independent of peripheral stimulation.
3. The effect of morphine hydrochloride on Anodonta has been investigated, and it is shown that, in addition to a possible narcotic effect, it abolishes the tonus of the unstriated portions of the adductor muscles.
4. A reflex is described which leads to the relaxation of the tonus in the adductor muscles, and the commencement of a new period of activity. The stimulus is vibration and/or rotation.
5. On the basis of the above conclusions regarding Anodonta a hypothesis is put forward to account for certain facts about the contraction and relaxation of the adductor muscles of Mytilus.