Throughout an animal's lifespan, synapses can adjust in remarkably dynamic ways in response to physiological fluctuations, but this dynamism may come at a cost to information transfer between pre- and postsynaptic cells and thus affect network stability. The Drosophila larval neuromuscular junction (NMJ) is a glutamatergic synapse that has been a premier model system for various aspects of synaptic biology, but its utility for more long-term studies is limited by the duration of the larval period, which normally takes just 4-5 days. Now, Dion Dickman and colleagues expand the horizons of NMJ research by terminally arresting larval development for up to 35 days. They accomplish this by knockdown of the gene encoding the transcription factor Smox in the prothoracic gland, which leads to a terminal larval stage during which larvae progressively grow and then decline, yet NMJ synaptic strength remains constant. Muscle growth during larval arrest is accompanied by a progressive expansion in the NMJ presynaptic compartment, matched by an expansion in postsynaptic receptor fields, with no difference in the size of synaptic vesicles. However, the probability of presynaptic release is decreased, explaining the constant strength of synaptic transmission and implying an adaptive adjustment to synapses during this novel life stage. Finally, the authors use arrested larvae to model neurodegeneration in stathmin mutants. Arresting larval development thus opens new windows to study the structure, plasticity, function and dysfunction of the fly NMJ, and also promises to be helpful to researchers with other interests in larval development or physiology.