PREHISTORIC JAPAN—BAELZ. 539 
to determine. I may remark in this connection that in China from 
the earlest times, the Emperor, the representative of heaven on 
earth, bore the title “The south looking Emperor.” The bodies 
were buried uncremated, but the bones at the time of the examination 
had usually disintegrated. Where the position of the body could 
be determined it was generally laid in the direction of the long axis 
of the structure, that is, north and south. The bodies lay on the 
floor, which was rarely paved with stones or covered with plaster, but 
at other times sar- 
cophagi of stone, 
terra cotta and 
wood were used. 
One dolmen usu- 
ally served for only 
one or two persons. 
Interment of a 
larger number was 
very infrequent and 
probably indicated 
a family vault or 
the death of many 
from some special 
occurrence. 
A particular form 
of grave is repre- 
sented by the im- 
perial graves (Jap. 
“Misasagi”) of the 
dolmen period. 
They would be 
more appropriately 
termed princely 
graves, since they FG. 6.—Japanese imperial grave (after Gowland). The length 
‘ < of the mound between water surfaces is 674 feet. The 
do not occur only outlines of the mound recall the schematic outlines of a 
in central Japan, ae figure. The dotted lines indicate rows of clay 
: cylinders. 
where the Emperor 
always lived, but also in all the districts where dolmens abound, and 
which must be considered as the seats of great feudal princes. These 
graves are often only a kind of unusually large dolmen mounds, yet 
they are prominent not only by their often enormous dimensions, but 
they have other peculiarities. In contrast to the position of the dol- 
mens on hills, these graves lie principally on plains. They are double 
mounds of a characteristic form (as the accompanying figure by 
Gowland shows), consisting of a trapezoidal mound flat on top and 
often terraced, joined to a higher circular one likewise flat on top. 
