The question of whether the synaptic extracellular matrix undergoes remodeling and how this remodeling is related to nerve terminal plasticity was examined in living neuromuscular junctions of adult frogs. Sartorius muscles were double stained with a fluorescent nerve terminal dye 4-(4-diethylamino-styryl)-N-methylpyridinium iodide (4-Di- 2-Asp) and rhodamine-tagged peanut agglutinin (PNA) which recognizes synaptic extracellular matrix. Both nerve terminals and synaptic extracellular matrix in 200 identified normal junctions were visualized in vivo two or three times over a period of 2.6–6 months. The majority of neuromuscular junctions (NMJs) showed remodeling of both nerve terminals and synaptic extracellular matrix. Only 2.5% showed no changes in either synaptic element. The most commonly seen remodeling involved correlated changes in both nerve terminals and synaptic extracellular matrix. In this large group, while some junctions (20%) showed overall proportionate changes in all branches, most junctions (68%) showed disproportionate extension and/or retraction of some but not all individual branches. Another group of NMJs (9.5%) showed mismatched changes in the nerve terminal and synaptic extracellular matrix. In this group, some NMJs showed a decrease in the nerve terminal length without a corresponding reduction in synaptic extracellular matrix length. In other junctions that displayed extension of branches, the PNA-stained matrix was longer than the distal tip of the nerve terminal. Morphometric analysis indicated an average increase of 15.6% in total nerve terminal length and 13.6% in total synaptic extracellular matrix length. Although almost all NMJs displayed remodeling in at least one branch, about 50% of the 2201 individual branches examined did not show changes. The average change was 8.9% growth in the length of individual nerve terminal branches and 8.3% growth in the length of individual branches of synaptic extracellular matrix. There was no significant difference in the morphometry between the repeatedly observed junctions and the previously unobserved control junctions. Furthermore, junctions in which the synaptic extracellular matrix was longer than the nerve terminal also were seen in control as well as in experimental muscles. Cases where the nerve terminals were longer than the synaptic extracellular matrix were never observed in newly arising junctional branches. The present study has shown extensive remodeling in not only the nerve terminal but also the synaptic extracellular matrix in adult living frog NMJs. Results suggest that nerve terminals retract before the synaptic extracellular matrix. A hypothesis that extension of synaptic extracellular matrix precedes nerve terminal growth during synaptic remodeling is proposed.