344 BUREAU OF AMERICAN ETHNOLOGY [Bull. 177 
pretations or to be overly cautious and limit conclusions to those 
clearly supported by the facts. Since only a small fraction of the 
content of a living culture is preserved in the best archeological record, 
a conservative attitude means that inferences will be minimal. If 
archeologists have a goal beyond pure recording, it is necessary for 
them to push the data as far as possible in the direction of interpreta- 
tion. From this point of view, the principal question is at what point 
this effort ceases to be scientific reasoning and becomes unwarranted 
speculation. 
The difficulty of arriving at criteria for distinguishing legitimate 
from unwarranted inference is illustrated by the evidence of trade re- 
lations between aboriginal groups in British Guiana. The ethno- 
graphic reports are consistent in describing extensive and intensive in- 
terchange of products. A quotation from Im Thurn will serve as an 
illustration : 
To interchange their manufactures the Indians make long journeys. The 
Wapianas visit the countries of the Tarumas and the Woyowais, carrying with 
them canoes, cotton hammocks, and now very frequently knives, beads, and other 
European goods; and, leaving their canoes and other merchandise, they walk 
back, carrying with them a supply of cassava-graters, and leading hunting dogs— 
all which things they have received in exchange for the things which they took. 
The Macusis visit the Wapiana settlements to obtain graters and dogs, for which 
they give ourali-poison and cotton hammocks; and they again carry such of these 
graters and dogs as they do not themselves require, together with more of their 
own ourali and of their cotton hammocks, to other Indians—to the Arecunas, 
who give in return balls of cotton or blow-pipes; or to the True Caribs, who pay 
in pottery. In this way, travellers with goods and with news constantly pass 
from district to district. [1883, p. 273.] 
When the archeological records of the Taruma Phase and Rupu- 
nuni Phase are examined for evidence of this contact, the results are 
very disappointing for an obvious reason: the items traded are pre- 
dominantly of a perishable nature whereas the archeological remains 
are confined to objects of pottery and stone. Only one Rupununi 
Phase site produced a few sherds of Taruma Phase vessels, and no 
Rupununi Phase items were found at any Taruma Phase site. No 
archeologist would feel justified in inferring from this slim evidence 
that communication between these two groups was close and continu- 
ous. A similar contrast exists between the ethnographically recorded 
frequency of contact between the Taruma and the Wai Wai and the 
absence of any archeological indication of contact with the possible 
exception of a dubious hint of acculturation in pottery decoration. 
On the other hand, the archeological record sometimes clearly sup- 
ports an inference of close and continued intercommunication. The 
large quantity of trade pottery of Mabaruma Phase origin in sites of 
the Abary Phase and the Koriabo Phase is an example. The con- 
clusion that these three Phases were contemporary and engaged in ex- 
