ON RUINS IN RHODESIA. 303 
so much care has been expended on the citadel, which is made of slightly 
dressed stones and ornamented with patterns on the walls. From Dhlo 
Dhlo there was obtained conclusive evidence of date. An outwork in 
front of the citadel had been used as a smelting place for tin, a metal 
which has not been discovered by prospecturs in Southern Rhodesia, but 
is known, on the other hand, to have been imported by the Portuguese. 
In the same part were found fragments of green glass and of Nankin 
china. From a great débris heap on the side of the citadel were recovered 
various objects, some of definitely medieval type, ¢.., a pair of iron 
manacles, which are represented on a sixteenth-century Valencia tile, 
glass beads (probably Indian), and Nankin china. There was no difference 
in charactér between the objects found in the upper and the lower levels 
of this kitchen midden. 
The absolute proof that these specimens belonged properly to the 
stone buildings in which they were found was furnished by the excavation 
of a hut-platform inside the entrance of the citadel, which was one of a 
series occupying the whole interior of the building and inseparably united 
with the stone walls of the front. Underneath the unbroken cement 
floor of this hut were found a sheaf of iron weapons with copper bangles, a 
spindlewhorl, fragments of glass and tin, ivory and glaze beads, and two 
large fragments of blue and white Nankin china with a flower pattern. 
This kind of Nankin china is known to be not earlier than the sixteenth 
century, and is therefore conclusive evidence for the dating of Dhlo Dhlo 
as actually post-medizval. 
The objects found at Dhlo Dhlo then fall into two main classes, viz., 
(1) implements, weapons, and ornaments of definitely African character, 
closely resembling what Kaffirs make at the present day ; (2) medixval 
imports traded by the Portuguese, chiefly from the Orient. 
And there is probably a third class, viz., metal work made by the natives 
in direct imitation of European processes, 
Everything that was observed at Nanatali, Khami, and Zimbabwe is in 
consonance with what had been discovered at Dhlo Dhlo. Each of these 
sites yielded the same mixture of definitely native objects with well known 
and unmistakable medieval imports. There was not a single specimen 
which could be described as ‘ancient’ in the archeologist’s acceptation of 
the word ‘The earliest possible dating for any object found at Zimbabwe 
is the fourteenth century a.p., and even that (Arabic glass) may quite 
possibly belong to the fifteenth century. The assertion that there is any 
superposition of periods at Zimbabwe is erroneous, a blunder on the part 
of an inexperienced excavator. There has not even been any rebuilding 
or restoration of walls ; the whole place is of a single epoch, and the date 
of that can be very closely fixed. 
Zimbabwe is a royal kraal which differs from other buildings in 
Rhodesia in no respect save for the greater massiveness of its walls and its 
more elaborate character. No single stone can have been placed there 
earlier than the fourteenth or fifteenth century a.p. Merely as an 
‘inhabited site the place may have been occupied a few generations longer, 
but not more. Such a hypothetical settlement would be prior to any of 
the buildings now standing at Zimbabwe, and would be contemporary 
with the northern sites, viz., Inyanga and the Niekerk Ruins, which the 
lecturer would ascribe to a century or two before 1500 a.p. 
From the character of the objects discovered, and from all the details 
of construction in the buildings and the dwellings they contained, there 
