According to Josephus, the Romans applied battering rams at three points along the wall (4, §20). How many breaches they actually made is quite another question. The surviving Greek manuscripts state “tîn ™reifqšntwn” literally “those [parts] that were torn down”. The translations, again, vary: breach (Bradshaw), broke through the wall (Thackeray), breaches (Williamson), and parts of the wall that were thrown down (Whiston). Further on, recounting the events after the first assault, Josephus states that ‘the bolder spirits guarded the gaps in the wall’ (7, §51), so we may assume that there was more than one breach, though how many exactly remains unclear.

Although Gutmann always referred to three breaches, as did Gichon (1987:79) the excavations show clearly only one: in a building below the synagogue . Here the wall was found broken down almost to its foundation and a huge number of arrowheads (some 300) and ballista balls (some 180) were found inside and outside of it. The wall is breached in a room of a domestic building. The original thickness of the wall was 0.70 m, and it was thickened to 2.05 m. by the addition of a secondary wall behind it, built mostly of fieldstones. Even at 2 meters, it was one of the weakest points in the wall; just south of it the wall is 4 m. thick. Did the Romans have information on the easiest point to breach the walls? As the room probably had no roof at this point in time, it may have been possible for someone standing on the cliffs above to judge the thickness of the wall. Or, the information may have been extracted from a prisoner, a fugitive, or — told by Josephus himself.

Another possible breach is located about midway along the wall where a small concentration of arrowheads was found in front of a 3.5-m. wide section where the wall was missing.

A third place is situated just above the synagogue, where an unusually carelessly patched-up passage through the wall was found, which might be a breach that was re-closed by the defenders after the first attack. Here not many arrowheads were found, but some 100 ballista balls tell of an artillery barrage.

Thus, there does not seem to be a correlation between missing sections of the wall and concentrations of ammunition, and the actual number of breaches will, unfortunately, remain conjectural.