Sec. in EXCAVATION OF THE ENDERE TEMPLE 421
SECTION II.-EXCAVATION OF THE ENDERE TEMPLE
On the morning of February 2I we continued south-eastwards, and after a march of about Approach
three miles sighted in the distance what the guides called the ` Potai ' of the ` Kane-shahr'. to Endere
Even from afar I could recognize through my glasses that it was a Stûpa, as I had, of course,
assumed when this feature of the site was first vaguely mentioned to me at the Mazâr. It was
interesting to note that on approaching the Stûpa we passed through a belt of sand-cones
thickly covered with living tamarisk and the scrub known as Ak-liken. It was here, at a dis-
tance of about three-quarters of a mile from the Stûpa, that we found an old well, which, when
cleared to a depth of 9 ft., yielded sufficient but rather brackish water. About half a mile
west of the Stûpa we reached eroded ground plentifully strewn with pottery fragments, and
this continued up to the ruin itself, which such typical ` Tati ' débris surrounded on all sides.
A rapid inspection of the Stûpa proved that the ground near it had been eroded to a depth
of from Jo to 15 ft. in different places, as clearly seen in Fig. 5o, which shows the ruin from
the south-west side, together with the bare loess and low sands behind it 1. Nor was I surprised
to find that the ruin had been dug into in two places, no doubt in search of treasure.
The contingent of labourers ordered up from Niya had arrived just when I was nearing 'Tati'
the Stûpa. Considering the great distance, some 120 miles, from which the men had been Stûua of
brought, and the difficulty of communicating with them over wholly uninhabited ground, I felt Endere.
not a little pleased at this well-managed concentration, which enabled me to start excavation
work at once. Leaving, therefore, the accurate survey of the Stûpa for later, I pushed on with
increased eagerness south-eastwards to where the remains of ` old houses ' were said to exist,
which held out better antiquarian promise than either Stûpa or the Tati around it. For about
a quarter of a mile the pottery-strewn ground continued, the eroded bare loess soil being exposed
in wide patches amidst dunes not exceeding 5 to 6 feet. Among the pottery fragments, mostly
small and showing long-continued abrasion, I noticed .a good deal of coarse black terra-cotta
and occasional pieces retaining a green glaze. Representative specimens of these materials
will be found described in the list of antiques (E. 006). Then followed dunes somewhat higher,
and among them occasional sand-cones with scanty tamarisk growth. Shrivelled trunks of dead
Toghraks emerged here and there from the drift-sand in the depressions between the dunes,
but no trace of structural remains appeared until I had arrived quite close to the dunes sur-
rounding the ruins which the guides from Imâm Ja`far Sadiq and Yârtunguz had spoken of as
` old houses '. The rows of wooden posts rising above the sand were a familiar sight. But
the high brick walls of some large building and the remnants of a massive clay rampart en-
circling the ruins presented a novel and very striking feature.
Going over the ground I soon realized that the extant portions of the rampart, which was First inspec-
largest and best preserved on the south, had belonged to a circumvallation approximately circular tien of
and enclosing an area which the subsequent careful survey (see plan in Plate XXXVI) proved to be nearly 420 ft. in diameter (including the thickness of the walls). The large building towards the eastern segment of the circumvallation (E. III) was found for the greater part almost clear of sand, and clear, too, of objects that could claim archaeological interest. But west of it a broad dune stretched across the interior of the fortified area, and near the centre of the latter I noticed rows of wooden posts just rising above the sand. Their arrangement in con-
' The line where the lowest course of masonry and the measuring staff; see also plan in PI. XXXVII.
original ground-level meet is marked by the foot of the to ft.