国立情報学研究所 - ディジタル・シルクロード・プロジェクト
『東洋文庫所蔵』貴重書デジタルアーカイブ

Thus on March 31, after the removal of bake-ovens a and b and the accompanying
child's grave, the red polished ware was still predominant among the "mixed"
pottery; but on the following day, when the excavation entered the layer of the
older pots, c and d, and the still deeper level of the second skeleton, the "mixed"
ware consisted principally of the painted group y, while there was a rare occur-
rence of the monochrome x. Very different, however, appeared the character of
the finds on April 5, when the layer of the older pots c and b had been removed
and the terrace had been sunk to between +27 and +25 feet. The monochrome
variety x was entirely lacking; there appeared only the group y. This corre-
sponded with the finds in the "galleries," where to a depth of +22 feet the only
fragments found were those of the pottery of group y.

It was already possible to draw the conclusion that the monochrome variety
x was the younger. Considering the manner of its occurrence, it was also probable
that it belonged to the same culture as that to which the walls, bake-ovens, kettles,
and skeleton of terrace I belonged, for the feeding-cup which was buried in the
child's grave is of the same kind as the great mass of monochrome pottery of the
upper layers. The question, to which period of house-construction the red mono-
chrome ware was to be referred, was completely solved by the discovery of the
quadrangles (see p. 87), for the pottery fragments with which the walls of these
cabinets were incrusted were of this characteristic ware. The only doubt possible
was in regard to the gray ware, which technically is far superior to the red. The
occurrence of the older varieties of vessels of group y in the upper layers is, however,
sufficiently explained when we consider that the inhabitants of these layers must
have materially disturbed the older layers, including their pottery, in establishing
their own dwellings. As a matter of fact, no pure layer of either group was found
in this transitional level of unmixed and monochrome ware, and such a pure layer
appears not to have existed.

Terraces IV and V (see figs. 28 and 29, plan and vertical sections of terraces
IV to VIII).—The finds in terraces IV and V correspond to the finds in terrace I.
Soon after the débris of the older excavations was removed and the undisturbed
earth of the hill was attacked, a child's skeleton (γ) was found in contracted or
Hocker position, and lying on its right side (cf. special report of Warner, No. o).
Its position is designated as in terrace IV. The horizontal layer of terrace V,
corresponding to the location of the exposed skeleton γ, has almost wholly disap-
peared, as the original surface of the hill there falls off abruptly to the south.
In terrace V there occurred first in situ a pot (e) of the kind found on the northern
half of the hill. To the east of this pot, and a little below the level on which
it stands, several skeletons were found lying in a heap (∂, No. o¹; cf. special report of
Warner); and farther east, rear these, were the remains of a deeper-lying pot (f)
in situ, of which the lip was destroyed (see fig. 30).

This stratiform succession of pots and skeletons resembles closely the finds
in terrace I. The levels are as follows: Skeleton, terrace IV +36 feet; lip of the
upper pot (e), terrace V, +34.5 feet; skeleton heap ∂ (cf. Mr. Warner's special
report No. o¹), +34 feet, near pot e; preserved edge of lower pot (f), +33 feet.