e management of the great Assyrians monarchs. Huge
dams seem to have been thrown across the Tigris in various places, one
of which (the Afrui) still remains, seriously impeding the navigation.
It is formed of large masses of squared stones, united together by
cramps of iron. Such artificial barriers were intended, not (as Strabo
believed) for the protection of the towns upon the river from a hostile
fleet, but to raise the level of the stream, in order that its water
might flow off into canals on one bank or the other, whence they could
be spread by means of minor channels over large tracts of territory. The
canals themselves have in most cases been gradually filled up. In one
instance, however, owing either to the peculiar nature of the soil or to
some unexplained cause, we are still able to trace the course of an
Assyrian work of this class and to observe the manner and principles of
its construction.

[Illustration: PLATE 134]

In the tract of land lying between the lower course of the Great Zab
River and the Tigris, in which was situated the important town of Calah
(now Nimrud), a tract which is partly alluvial, but more generally of
secondary formation, hard gravel, sandstone, or conglomerate, are the
remains of a canal undoubtedly Assyrian, which was carried for a
distance of more than five-and-twenty miles from a point on the Khazr or
Ghazr Su, a tributary of the Zab, to the south-eastern corner of the
Nimrud ruins. [PLATE CXXXIV., Fig. 1.] Originally the canal seems to
have been derived from the Zab itself, the water of which was drawn off,
on its northern bank, through a short tunnel--the modern Negoub--and
then conducted along a cutting, first by the side of the Zab, and
afterwards in a tortuous course across the undulating plain, into the
ravine formed by the Shor-Derreh torrent. The Zab, when this part of the
work was constructed, ran deep along its northern bank, and, sending a
portion of its waters into the tunnel, maintained a constant stream in
the canal. But after awhile the river abandoned its north bank for the
opposite shore; and, water ceasing to flow through the Negoub tunnel, it
became necessary to obtain it in some other way. Accordingly the canal
was extended northwards, partly by cutting and partly by tunnelling to
the Ghazr Su at about two miles above its mouth, and a permanent supply
was thenceforth obtained from that stream. The work may have been
intended in part to supply Calah with mount