Excavations in the third trench on the north side of the "gymnasium" were halted when we uncovered a layer of eroded material (colluvium) in the southern part of the trench. The material included some traces of volcanic ash which was washed away. This deposit undoubtedly dates to the period before the development of the area and the construction of the complex, as it contained no ceramic material that would have been dated later than the second century A.D. The geological team led by P. Muchez identified the subsoil, on top of which all north walls of the complex were built, as weathered volcanic tephra. It probably originated from the nearby volcano of Gölcük (ca 10 km north of Sagalassos), rather than as a weathered ophiolitic melange as previously suggested. As in the two previous trenches, no floor level was exposed in this trench, although the floor of the space preceding the large central room (either a portico or courtyard) must have been located on top of this deposit. The presence of a very compact layer containing second-century A.D. material on top of it, which can most probably be identified as the floor substrate, seems to confirm the floor's location. The actual floor seems again to have been completely removed. No wall or other type of division was retrieved between the central room and this space. However, the fact that the supposed floor substrate in the southern part of the trench is some 0.90 m below that of the central room suggests a kind of separation between the two. Above the supposed floor substrate we removed a total of five layers of debris of both occupation as well as construction origin. Total thickness was 2.10 m, and the layers contained archaeological material dating between the late fourth and the early sixth century A.D. The debris layers confirm the observations made in trench 2 that the site was used as a dumping ground for several decades after its abandonment in the late fourth-early fifth century A.D. It is also striking that the coins retrieved from the dumped material, nearly 100 of them, all belong to the second half of the fourth century A.D. Most recent numismatic evidence consists of coins minted by Arcadius around the transition to the fifth century. All evidence suggests that the "gymnasium" in fact was torn down during his reign.

We laid out a fourth trench in the southwest corner of the complex, this in order to expose the south wall of the southern row of rooms. Here, even the wall itself was completely dismantled, leaving only its foundation, composed of mortared rubble, standing. The method of construction is similar to that of the foundations of the northern walls. Unfortunately, we could not confirm the date of origin by archaeological material since none was retrieved from the layer in which the foundation wall was built. Judging from the ceramic assemblage retrieved from the layers of architectural debris above the floor substrate, the south wall also appears to have been dismantled in the late fourth-early fifth century A.D.

A final trench was laid out 7 m to the west of trench 4 in order to investigate the western boundary wall of the partially curved elongated structure that preceded the building to the south. This structure was tentatively identified as a running track on the basis of the results of the geophysical survey (see Near the Theater, July 11-15). The curved wall at the western extremity of this structure proved to be a line of rubble stones rather than an actual wall. This series of stones appears to have been put to divide two types of fill. To the east of the wall a layer including large limestone blocks was excavated, while to the west the deposit contained a huge amount of medium-sized rubble. This rubble may indicate that construction debris was disposed off of there, although no traces of mortar remain. Unlike the northern trenches, we retrieved hardly any artifacts, which indicates that unlike the northern part, the south part of the complex was not used as a dumping ground for occupation waste. Since the uncovered structure cannot be related to the gymnasium, the excavation in this trench was halted. Although the nature of the terrain does not exclude it, we found no indications for a running track.

At the end of excavations near the theater, we can only conclude that the geophysical map of the building, confirmed by the subsequent test soundings, corroborates the identification of this large building as a gymnasium. It must have been built during the middle of the second century A.D., probably in connection with the construction of the nearby theater. As was also the case with that other monument connected with the pagan Greek education, i.e. the Neon-Library, the monument was completely dismantled around the transition from the fourth to the fifth century A.D. As if to eradicate these pagan symbol forever, it was stripped of all its wall and floor coverings. Most of its walls were taken down to a level just above or even to their foundations proper. The north part of the complex was used for at least another two centuries as a dumping area.