inhabitants fled into the Mesopotamian desert, while the remainder
sought refuge in Tela at the foot of the Ukhira.**

* The site of Mariru is unknown; according to the text of
the Annals, it ought to lie near Severek (Kinabu) to the
south-east, since after having mentioned it, Assur-nazir-pal
speaks of the people of Nirbu whom he engaged in the desert
before marching against Tela.

** Tila or Tela is the Tela Antoninopolis of the writers of
the Roman period and the present Veranshehr. The district of
Nirbu, of which it was the capital, lay on the southern
slope of the Karajah-dagh at the foot of Mount Urkhira, the
central group of the range. The name Kashiari is applied to
the whole mountain group which separates the basins of the
Tigris and Euphrates to the south and south-west.

The latter place was a strong one, being surrounded by three enclosing
walls, and it offered an obstinate resistance. Notwithstanding this, it
at length fell, after having lost three thousand of its defenders:--some
of its garrison were condemned to the stake, some had their hands,
noses, or ears cut off, others were deprived of sight, flayed alive,
or impaled amid the smoking ruins. This being deemed insufficient
punishment, the conqueror degraded the place from its rank of chief
town, transferring this, together with its other privileges, to a
neighbouring city, Tushkhan, which had belonged to the Assyrians from
the beginning of their conquests.* The king enlarged the place, added to
it a strong enclosing wall, and installed within it the survivors of the
older colonists who had been dispersed by the war, the majority of whom
had taken refuge in Shupria.**

* From this passage we learn that Tushkhan, also called
Tushkha, was situated on the border of Nirbu, while from
another passage in the campaign of the Vth year we find that
it was on the right bank of the Tigris. Following H.
Rawlinson, I place it at Kurkh, near the Tigris, to the east
of Diarbekir. The existence in that locality of an
inscription of Assur-nazir-pal appears to prove the
correctness of this identification; we are aware, in fact,
of the particular favour in which this prince held Tushkhan,
for he speaks with pride of the buildings with which he
embellished it. Hommel, however, identifies Kurkh with the
town of Matiato, of which mention is made furth