eat interest to me.

There are a great many theories regarding these former salt lakes, and it
is not easy to say which is right and which is wrong. The general belief
is that these lakes were formed by the overflow of the Halmund swamp into
the Shela (river) which carried sufficient water not only to fill up the
God-i-Zirreh, but to overflow when this was full into the next depression
east of the Zirreh.

There is no doubt that to a great extent this was the case, but these
lakes were, I think, also fed more directly by several small streams
descending from the mountains to the south and west of the Zirreh, which
form the watershed--and very probably also from the north by the Halmund
River itself. Both lakes were dry and seemed to have been so for some
time. The God-i-Zirreh, forming now a great expanse of solid salt some 26
miles long by 5 or 6 wide, extends in a long oval from west to east. The
other lake was somewhat smaller.

To the south of these salt deposits in the zones between them and the
present Afghan boundary, and forming the southern fringe of the Afghan
desert, the soil is covered with gravel and stones washed down from the
mountain sides. Very stony indeed is the desert towards the Malek-Siah
end, then further north-east appear brown earth, shale, and sand. To the
north of the lakes was a long line of bright yellow sand extending from
west to east and broad enough towards the north to reach the bank of the
river Halmund. Another shiny patch, which at first, from a distance, I
had mistaken for another smaller lake, turned out on examination to be a
stretch of polished shale which shone in the sun, and appeared like
bluish water.

Stunted tamarisk grows in some parts but not in the immediate
neighbourhood of the salt deposits. We have here instead a belt of
myriads of small conical sand-hills, also spreading from west to east,
quite low to the west and getting higher for several miles towards the
east. In the south-west part of the desert, curiously enough, between the
zone of conical hills and the salt deposits, and parallel to both, lies a
row of semi-spherical sand and salt mounds of a whitish colour.

To the east-south-east of the lakes the sand-hills rise to a great height
and eventually form a high ridge, which for some reason or other is cut
perpendicularly on its western side, possibly as the result of a volcanic
commotion. Of similar origin probably was the gigantic crack caused by an
earth