National Institute of Informatics - Digital Silk Road Project
Digital Archive of Toyo Bunko Rare Books

Then in t years from the time that a decrease in fall began, S = P − G − pt.
If P in the beginning had been equal to 2G, and thus evenly divided between
surface and underground drainage, and p equal to say P/2000 = G/1000,

S = G − G/1000 t,

and total fall (S + G) = P − P/2000 t.

Plotting these sloping lines we see how a steady decrease in precipitation, as
the underground value is approached, affects the surface drainage during the first
century by only 10 per cent, during the sixth century by 20 per cent, during the
ninth century by 50 per cent, and during the tenth century by 100 per cent; that
is, a uniform decrease subtracts an increasing proportion of what is left till the
surface excess vanishes, though there may still be a plentiful fall, which is there-
after consumed by underground drainage. Whatever record remained would
thus have the appearance of sudden change towards the end.

We are thus driven to believe there has been a relatively recent decrease
in the precipitation. The next question is: Where does the underground water
drain to? A small proportion finds its way to the surface again, supplying most
of the present flow in trunk-streams, a return through small cavernous springs,
several of which were observed near Manisht. But the limited amount which
now composes the Anau Su and Gyourse Su, even at high water, represents but
a slight proportion of what falls over that high region. That it continues out
under the plain is evidenced by water rising in native wells scattered far and wide
over the Turkoman Trough. It seems barely possible that water drains through
sand and gravel beds all the way from these mountains to the Caspian.

Turning to the delta again, we find it tells the same story in a recent shrinkage
of water distribution. Wide areas of clay, some of which project 5 to 10 miles out
among the dunes, are now dry during the highest floods, and not many centuries
can have elapsed since their alluviation, for all would otherwise have been strewn
with dunes. Some few hundred years ago water found its way to a small town
near Ball Kuwi, about 6 miles north, whose ruins we explored.

SHAPE OF THE ANAU DELTA AND IRREGULARITIES WROUGHT BY MAN'S DÉBRIS OF
OCCUPATION AND HIS CONTROL OF ALLUVIAL DEPOSITIONS.

The Anau delta and its surroundings present a myriad of surface problems,
of whose most general features a competent study would involve a large-scaled
topographic map, constructed with the double eye of physiography and arche-
ology; while far more light on both past and present may be revealed with the aid
of a microscope and sensitive weighing balance. When for the first time one stands
upon its clay surface, all the plain appears in simple flatness to the eye. Rising
to the south is the Kopet Dagh Range of rather flat-backed mountains, notched
by valley gaps, and from their base the plain slopes gently out into the northern
semicircle of ocean-like horizon, broken only by sand waves tossed from the desert
of Kara Kum. Toward the middle rise two kurgans and the last citadel of ruins
with its outlying watch-towers, the three dead towns of Anau in whose stratified
débris we dug.