depth and one at 8 m below floor level, while, in the living
memory of local informers, the water table used to be found at 15
m or more. The fact can be of importance for the reconstruction of
ancient climatic conditions, even if the debit was always slow.
The long history of the house - over six centuries of
continuous use - cannot be followed progressively, as floor levels
remained stable throughout: cemented floors, in some rooms still
smooth and painted red, in some much wom, were made and
remade without aceumulation layers in between. We present the
plan of the last stage, shortly before the frnal abandonment in the
9 th century, but it should be kept in mind that some modifications
of the original plan were noticeably earlier (Fig. 2). According to
the evidence collected this year, the house survived in its more or
less initial stage until the late 6 th century, when an earthquake
seriously damaged some walls, including the partition wall between
the two main areas of the house and their twin entrance. As a
result, both doors were blocked, the former guest entrance through
loc. 17 became dependent on the other courtyard and, as both sets
of stairs in the family wing were dismantled, the upper floor seems
to have disappeared from use in this part of the building.
During the Islamic period the porticoes of the family wing
were closed and the original apartments became three neatly
separated households sharing a common courtyard. The eastem
part of the courtyard, which remained open until then, was built
over with two rooms, each provided with a door leading from the
street. One (loc. 14) became the main entrance, the other (loc. 15)
was a stable equipped with a manger. In the meantime the
reception suite was transformed into two more households, each
complete with a yard.
A typical feature of this late stage are cupboards hollowed out
in walls at different levels. There are also tannur ovens and storage
jars set into floors. Many doors were blocked, isolating formerly
interconnected rooms. In spite of these innovations, which reflect
71
memory of local informers, the water table used to be found at 15
m or more. The fact can be of importance for the reconstruction of
ancient climatic conditions, even if the debit was always slow.
The long history of the house - over six centuries of
continuous use - cannot be followed progressively, as floor levels
remained stable throughout: cemented floors, in some rooms still
smooth and painted red, in some much wom, were made and
remade without aceumulation layers in between. We present the
plan of the last stage, shortly before the frnal abandonment in the
9 th century, but it should be kept in mind that some modifications
of the original plan were noticeably earlier (Fig. 2). According to
the evidence collected this year, the house survived in its more or
less initial stage until the late 6 th century, when an earthquake
seriously damaged some walls, including the partition wall between
the two main areas of the house and their twin entrance. As a
result, both doors were blocked, the former guest entrance through
loc. 17 became dependent on the other courtyard and, as both sets
of stairs in the family wing were dismantled, the upper floor seems
to have disappeared from use in this part of the building.
During the Islamic period the porticoes of the family wing
were closed and the original apartments became three neatly
separated households sharing a common courtyard. The eastem
part of the courtyard, which remained open until then, was built
over with two rooms, each provided with a door leading from the
street. One (loc. 14) became the main entrance, the other (loc. 15)
was a stable equipped with a manger. In the meantime the
reception suite was transformed into two more households, each
complete with a yard.
A typical feature of this late stage are cupboards hollowed out
in walls at different levels. There are also tannur ovens and storage
jars set into floors. Many doors were blocked, isolating formerly
interconnected rooms. In spite of these innovations, which reflect
71