Since synaptic plasticity is an important property of the brain, it is timely to try to understand the possible mechanisms underlying this phenomenon. The role of the cytoplasm for neuronal functions has until now been largely overlooked, the main emphases being on the plasma membrane for fast electrical events and on cytoplasmic organelles for the slower metabolic processes. However, recent studies on the cytoplasm of non-muscle cells have stressed the importance of contractile proteins, like actin, on maintaining the cell shape and a number of vital cellular functions, which may be related to the phase transitions in the cytoplasm. The necessary versatility is conferred on the actin networks by actin-associated proteins and by the free cytosolic calcium. In the nervous system, in addition to actin and myosin, a number of actin regulatory proteins was recently isolated, and they were shown to have properties similar to those of other non-muscle cells. Consequently, actin networks in neurons like those in non-muscle cells may be capable of contraction and phase transitions. The phase transitions have a rapid onset, and they may be quickly terminated or they may last over extended periods of time. In this way actin networks may gain control over the state of the cytoplasm and hence over the function of the neuron. Actin may be, therefore, uniquely suited to regulate various plastic reactions. The cytoplasm of growth cones and dendritic spines contains solely actin networks and is devoid of microtubules and neurofilaments. Since both these structures contain myosin and since growth cones are endowed with a considerable motility, dendritic spines also may have a likewise property. The necessary regulation of the levels of free cytosolic calcium may be provided by the spine apparatus in addition to calcium pumps in the plasma membrane and calcium regulatory proteins in the spine cytoplasm. Various types of stimulation which change the level of free cytosolic calcium may induce contraction of the spine actin network which may be responsible for the morphometric changes observed following different experimental interventions and pathological conditions. Although most of the conclusions in this review are rather speculative, they may provide directions for future research in the spine and synaptic plasticity.