Some Description of the Rio Los Amigos, Madre de Dios, Peru Robin B. Foster Environmental & Conservation Programs The Field Museum, Chicago 20 April, 2001 The principal drainage of the Southwestern Amazon Basin is the Madre de Dios/Madeira river system. In Southeastern Peru the Madre de Dios and its extension, Rio Manu, come up to and drain the eastern slopes of the Andes. But they also drain the relatively flat Amazon Plain to the north. One of these northern tributaries is the Rio Los Amigos. Until it joins the big river, the Rio Los Amigos and its branch the Amigillos run roughly parallel to the Rio Madre de Dios. Its mouth is half-way between the junction of the Manu and the junction of the Rio Piedras (another northern tributary, also roughly parallel to the Amigos). The Rio Los Amigos passes through two very different kinds of Amazon terra firme. From its mouth north to the split of the Amigillos and for 15 km further upstream, the Amigos passes through an area of high terraces, notable for being very flat with only infrequent cutting by small streams. These flat terraces are the very western tip of a formation that forms a broad regional arc of weakly dissected uplands. This formation does not go south of the Rio Madre de Dios except in the vicinity of Puerto Maldonado (where it crosses over to just beyond the Rio Tambopata), but instead sweeps northeast into Pando, Bolivia, eastern Acre, Brazil, and beyond. The vegetation of these flat terraces has a high (40 m) mostly-closed canopy. It is characterized by a high density of castanas (Berthlletia excelsa) and other emergent trees of the same family, Lecythidaceae. These are of course mixed with hundreds of other tree species, but few as prominent. The trees are mostly straight-trunked with relatively small crowns, stranglers are rare, density of lianas is relatively low, and herbs, epiphytes, and trunk climbing plants are few. For the most part it is a beautiful and easy-to-walk-through forest, and in this western end of the formation is remarkably undisturbed, regardless of the obvious visits by castaneros. Nor does it show any of the signs of having been extensively cleared several hundred years ago, such as are found on the hills and terraces in much of Pando near the Rio Tahumanu. Where this terra firme has been disturbed in the past (other than by downburst windstorms) is along stretches of the bluffs over the Amigos floodplain, presumably by indigenous peoples over hundreds of years. These areas are now thick with bamboo (pacales) in well defined blocks along the bluffs such as northeast of the Centro Rio Amigos station (near where the floodplain of the Amigos meets the floodplain of the Madre de Dios) and at the bifurcation of the Amigillos and Amigos. The second kind of terra firme is encountered about 40 km straight up the Amigos from its mouth and 60 km following up the Amigillos. Here, sometimes abruptly and sometimes gradually, there is a transition to highly dissected steep hills ~50-100 m high or higher. This is the southernmost end of a large regional physiographic formation, interrupted only by rivers, that stretches northwest and north for hundreds of kilometers in to the Ucayali Department of Peru, western Acre, Brazil, and beyond. It also does not pass south of the Rio Madre de Dios, though it does appear to be on both sides of the Manu floodplain above the Rio Pinquen. All the upper reaches of the Amigos and Amigillos drain from this formation. The vegetation of the dissected hills occupies the largest area of the Madre de Dios Department and is the least known. Much of it is also not particularly inviting. Large parts of the area are covered with an understory of spiny bamboo (three species of Guadua), mostly under a sparse tree canopy but occasionally as open solid stands. Other large parts are covered with dense vine tangles. Yet other parts seem to have closed canopy forest. While perhaps not as attractive as the flat terrace forest, the mystery of the dynamics and history of these different vegetation types is an intriguing, challenging, and important problem. Why these two physiographic formation? Perhaps the flat terraces are geologically younger. Perhaps the dissected hills are being raised faster from below by the upthrust from the Nazca plate sliding under the continent. Perhaps the composition of these ancient sediments are different, resisting erosion in the flat terraces, succumbing to erosion in the dissected hills. The third important formation is the Amigos floodplain itself. Although the river is a meandering one, the formation of oxbow lakes (cochas) is either not very common (except near the mouth) or they fill rapidly. Small stands (aguajales) of swamp palms (Mauritia flexuosa) are frequent along the margins of the floodplains. Succession on the meander beaches appears to be similar in composition to that of the Manu and Madre de Dios meanders, though perhaps not as rich in species, but a smaller version with the same process of forest formation. The easy access to mostly intact versions of the two major terra firme formations of southeastern Peru, the unspoiled floodplains of the Amigos and Amigillos, not to mention the kaliedoscopicarray of barely-studied floodplain habitats and low terraces along the adjacent Rio Madre de Dios and south of it, all argue strongly for this area as an ideal center for both basic research and for research on land and forest management of southeastern Peru in particular and the southwestern Amazon in general.