326 BUREAU OF AMERICAN ETHNOLOGY [Bull. 177 
of occupancy and a high frequency of movement of the village. The 
majority of the sites are on the savanna, usually occupying a knoll or 
hilltop and often bordered on one side by a patch of trees. A few 
sites in the earlier part of the sequence are on.a forested slope not 
more than a kilometer in from the edge of the savanna. Proximity 
to rivers or even large streams is not characteristic, implying that 
watercourses did not furnish the major avenue of transportation and 
communication. 
The seriation of Rupununi Phase aluen lates sites is based. on an 
increasing frequency of Rupununi Plain and a corresponding de- 
creasing frag pene of Kanuku Plain. The trends are such that the 
sites in the lower half of the sequence havea majority of Kanuku 
Plain and those in the upper half have a, majority of Rupununi 
Plain. None of the types of decoration has any temporal or quan- 
titative significance, and the variety of techniques seems’ to reflect 
outside influences felt at different times by certain villages. 
The most distinctive stone tool associated with the Ruins Phase 
is a stubby, rectanguloid ax with lateral notches (fig. 119; pl. 61). 
More abundant than axes, but restricted to the upper part of the 
sequence, are stone hoes (fig. 121; pl. 62) made of flat slabs of rock 
roughly shaped by percussion blows and with a high polish -on .the 
tip of the blade produced by digging in the sandy soil. The appear- 
ance of these tools only after atic European contact suggests 
that they are an imitation of a European metal hoe deemed superior 
to the aboriginal digging stick. Other stone artifacts include bowls, 
choppers, hammerstones, manos, and metates. 
The Rupununi Phase can be equated with the Macushi and Wapi- 
siana Indians who are known to have occupied some of the later sites 
in the sequence. The sampling of sites represented in the Rupununi 
savanna survey leads to the conclusion that the entry of these groups 
onto the savanna is very recent. Present evidence makes it impos- 
sible to date the intrusion earlier than the end of the 18th century. 
Use of the savanna for hunting at an earlier time is probable, but the 
choice of the savanna for residence is clearly a post-European phe- 
nomenon, dictated by the economic advantages of proximity to Euro- 
pean settlements. 
COMPARATIVE DATA, CONCLUSIONS, AND INTERPRETATIONS 
The archeological picture on the Rupununi savanna is of a very 
recent occupation by a single pottery-making group, designated as 
the Rupununi Phase. Prior to the arrival of this group, the region 
was probably exploited by wandering hunters and gatherers, but 
evidence of their existence is restricted to a few scattered’ finds of 
chipped stone tools. Sites of the Rupununi Phase include habita- 
