486 ANXUAL REPORT SMITHSONIAN INSTITUTION, 19 30 



moat was subsequently found at two other widely separated points 

 where we made sections through the extramural debris, and must 

 be regarded as a well conceived part of the city's defences. 



In the course of the summer we cleared a crescent of more than 

 a hundred meters of the wall, with adjacent city areas, but no trace 

 of a gate was found at the north end. However, in view of the size 

 of the city, it seems certain that there must have been another gate 

 besides the one at the south end. Two stretches of city wall still 

 remain to be excavated, one on the west, the other on the east side 

 of the Tell. Since, anciently, the north-south road passed on the 

 west side of the Tell, the probabilities are in favor of a second gate 

 on the west side. This is a problem that still awaits solution. One 

 more season will enable us to complete the excavation of the entire 

 mound and so find the answer for this as for a number of other 

 questions. 



NEW FEATURES OF THE CITY WALL 



The attack upon the Tell from the north end has had the advan- 

 tage of revealing the fact that the city's defenses have had a more 

 complicated history of construction than appeared from the earlier 

 excavations at the south end. For instance, the inner and older city 

 wall was not found at the north end, and hence no intramural area, 

 filled in its upper level with grain bins. The main wall itself showed 

 striking differences of construction. The builders had first exca- 

 vated a wide trench, carried to bedrock, and this they filled to a 

 height of 2 meters with loose rocks, mostly small. Upon this bed 

 of rock fill the wall was built with courses of large stones, laid with 

 clay mortar. The steady and increasing pressure of accumulating 

 debris against the inner face of the city wall had gradually pushed 

 it out so that, in spite of its great thickness, it began to lean outward 

 more and more. This action was facilitated by the loose founda- 

 tions and the absence of all counter pressure against the outer face 

 of the wall, which coincided with the edge of the sharply descending 

 rock slope of the hill. Ultimately it became necessary to save the 

 wall from total collapse by building a buttress wall along the outside. 



The structural peculiarities of this north-end wall made it neces- 

 sary to ascertain, if possible, whether it was built at the same time 

 as the Bronze Age wall at the south end. I had two sections cut 

 through it, and the potsherds found in the masonry were carefully 

 collected in baskets and labeled according to the successive courses 

 of stones. There were numerous Early Bronze Age fragments, but 

 a considerable proportion of the potsherds belonged to the Iron Age, 

 a decisive indication that the wall, at the section points, could not 

 have been built earlier than the Iron Age. The lowest course of 

 the wall yielded only Early Bronze Age pottery, but this may have 



