From the first glimpse of moving spines half a decade ago, the prevailing view has been that spine contortions and wiggling, especially during development, maximize encounters with presynaptic growth cones or synaptic boutons. Other new evidence has revealed that spines continue to be motile even after they settle on a presynaptic partner and form a synapse. We present the evidence for each view, and discuss how spines with synapses could move relative to their apparently stable presynaptic partners. Thus, spine motility might not simply be a means towards an end of synapse formation, but could continue, albeit at a lower rate, during synapse turnover after development ends.