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

Pithos *a* has, indeed, lost its lip, but the height of what remains is 33.3 cm.; hence,
the height of the floor of period III can be estimated at about +26 feet or a little
higher. The position of the brick does not suffice to determine its height, since
it may have fallen from above into débris. The clay chest suffices to determine
the elevation of the floor of period II. As long as it was in use, together with
its corresponding pithoi *b* and *c*, the floor can not have been higher than the pre-
served edge, *i. e.*, +25.5 feet; indeed, it probably lay somewhat deeper, since
pithos *b* (+23 feet), after being put together, shows a height of 65 cm. Pithos *c*
at 23.7 feet has only 62 cm. height. The floor of the clay-wall layer is also unde-
termined. It probably lay deeper than its preserved upper edge (+20 feet),
since the inner sides of the thick walls were covered with a layer of white lime
and must have been still visible. Therefore, the skeletons can not be brought
into connection with the oldest period (I). They all lie above the presumable
level of the floor and must, therefore, be younger.
If now we consider the skeletons from the bottom upwards, *δ*, *ζ*, *γ*, and *ι*
can be brought together, since they all lie between +23 and +24 feet. This
position corresponds to period II. From this period are excluded skeletons *γ*,
*ε*, *ϑ*, since they lie at +25 feet. We must, therefore, assign them to the following
period (III), of which the level lies at +26 feet or a little higher. There remain,
therefore, for period IV only the two skeletons *α* and *β* at +27 feet. Here also
it is to be assumed that the burials were very shallow—immediately under the
floor—an assumption which is justified by our experience in the North Kurgan.
If this were not so the skeleton groups would each have to be shoved upward by
one period; that is, skeletons *δ*, *ζ*, *γ*, and *ι* would have to be assigned to period
III, *γ*, *ε*, *ϑ* to period IV, and *α* as well as *β* to a still younger period, for which
no determining finds have as yet been made in terrace B.
The buildings of period I seem to have been destroyed by fire, for the débris
above it is much mixed with charcoal, and remains of burnt clay. It is possible
that in the construction much wood was used. The layers of débris cover the
remains of the walls and form the basis for the clay chest, pithoi, and skeletons.
Skeleton *ι* lies upon exactly such a burnt layer of the lower house-complex.
Figs. 48 and 49 will serve to show the relation of these layers. They repre-
sent the longitudinal sides of terrace B at the close of the excavation. A still
higher wall is standing in the upper part of the west side of the terrace, between
+32 feet and +31 feet; it must, therefore, be younger than the threshold. At
about the height of its lower edge there was found a well-preserved dagger-blade
or lance-point of copper. At the same level, outside of the terrace and immedi-
ately below the surface of the hill, there was found a well-preserved and almost
perfect vessel of group 5 (plate 10, fig. 1). This find shows how the horizontal
layers on the declivity come to the surface, through the gradually progressing
deformation of the hill and there fall a prey to disintegrating forces.
Lastly, we must mention the principal finds.
As regards the pottery, it has already been remarked that groups 3 to 5 are
represented until the level of +28.25 feet is reached. Below this, the ware of
greenish clay, group 5, is conspicuously predominant. The red ware is really not