‘GAZETTEER OF PERSIA VOL. I Comprising the Provinces of ASTARÁBÁD, SHÁHRUD-BÚSTAN, KHÚRÁSÁN, AND SÍSTÁN’ [‎204r] (414/722)

of ten inches, taking mountain and valley together. Throughout the
greater part of central and south-eastern Persia and Baluchistan, the
annual rainfall cannot be much more than five inches. On most parts of
the plateau the rainfall is divided as follows : A little rain is hoped for,
but not always expected, in November, to sow the early crops. In Decem
ber there is generally a tolerably heavy fall of snow, and another in Febru
ary ; followed by showers in March and the beginning of April; after which
there is nothing but an occasional thunderstorm in the mountains till the
next winter. ere it not that the lofty hills store the moisture in the
shape of snow, nine-tenths of Persia would be the arid desert that half of
it is now. As it is, cultivation over the greater part of the country
is only possible by artificial irrigation, either by canals or by the system
of wells connected by underground channels called kanat or karfz, and
peculiar, I believe, to the Iranian plateau.
With so scanty a rainfall, great rivers are not to be expected ; and we
accordingly fail to find a single stream unfordable during the dry months
throughout the whole of the interior plateau. The Caspian watershed,
besides the Kur and the A raxes, presents three considerable streams in
Persian territory. The first is the Kizil-Uzan or Saffd Rud, which drains
about 25,000 square miles of country east and south of the Urmia basin.
On the south-west of the Caspian the streams, though often of considerable
volume, are of small length. The most important is the Lar, which rises
north of mount Damavand, and flows round its southern foot, thus nearly
encircling it before completing its course to the sea.
The other two are the Atrak and Gurgan, of which the former and
larger is very incorrectly laid down on existing maps of Persia, being
shown as rising near Abivard, about long. 60°, flowing due west through
four degrees of longitude before turning south-west to the Caspian, and
receiving many affluents from either side in its course.
Russian explorations and the travels of Colonel Valentine Baker and
Lieutenant Gill, r.e., have largely modified this view. It appears that
the true Atrak is the same stream as that shown on maps as the “ Germch
Khaneh^ or “ Germe Rud/'’ This has its origin in a dry torrent near Kabu-
shan or Kushan, and flows through Shfrvan and Bujnurd in a north-westerly
direction, receiving several considerable streams from the Kuran Dagh on
the north, but none of importance from the hills to the south. As the
Atrak river is supposed in certain quarters to form the frontier of Persia
throughout its course, this modification is one of considerable importance,
as it places the Persian town of Shfrvan and the flourishing Persian district
of Daragaz outside the limits of the kingdom. In describing the present
state of the political frontier of Persia at the close of this chapter, I shall
advert to this subject again.
The streams draining southern and western Persia into the sea dimin
ish regularly in importance from north-west to south-east. The slopes
of the mountains of Kurdistan and the Zagros pour their waters into the
great river Tigris through the Diala and Karkhah rivers. Further south
the Diz and Karun, or Kuran, unite their waters in the plain of Khuzistan,
to form a river navigable from the sea to the first range of hills. It is
worthy of remark that this stream, which formerly discharged its whole
volume direct into the sea, now parts with the larger half through an
artificial channel into the Shatt-ul-Arab, the common outlet of the Tigris
46