While the preceding statement is true of numerous areas in New World archaeology, the dual problems of low surface visibility and limited accessibility are probably nowhere greater than in the tropical lowlands of South America, particularly Amazonia. These problems, combined with the sheer immensity of the area and small number of archaeologists working there, have led to widely conflicting interpretations of the archaeological record, not to mention opposing reconstructions of macro-regional prehistories (Gibbons 1990; Roosevelt 1991). In spite of these differences, all archaeologists working in this area have had to confront the severe logistical constraints and preservation biases imposed by the humid tropical environment, and in so doing, have been forced to employ “methodologically unlovely techniques” of one kind or another.

My purpose in this chapter is not to enter into this protracted debate on Amazonian prehistory. Instead, I explore certain methodological themes relating to archaeological survey and site discovery in the forested tropics of lowland South America, using as my point of departure Donald Lathrap's seminal article entitled “Aboriginal Occupation and Changes in River Channel on the Central Ucayali, Peru” (1968a).