Bluntschli, however, did not make any chemical analysis of water and evidently did not know that Katzer had already discovered the surprising chemical pureness and weakness of most Amazonian waters (Katzer, 1903). Otherwise, that keen observer would have discovered deeper and more revealing interrelationships. Systematic research on Amazonian water chemistry started only after 1945 and still continues. It confirmed Katzer's early discovery and extended it to by far the largest areas of the whole forest-covered Amazonian region: the 'hylaea' of Alexander von Humboldt (Sioli, 1950, 1954, 1955; Schmidt, 1972; Furch, 1976; Junk and Furch, 1980). The only exceptions to that general rule are a few small local zones with peculiar geology and lithology (Sioli, 1963). It is especially true of the strip of land of the Andean foothills with its projections along the whitewater rivers and the alluvial lands of the vdrzeas. Here we find chemically richer waters, signs of richer, more fertile soils, and conditions completely different from those of the terra firme. Most Amazonian waters, however, prove to be so pure and poor in electrolytic content that they are best compared to distilled water of low quality and almost equal to rainwater. In a climate as humid as that of the Amazonian hylaea, and under a mature forest cover (i.e., with a biomass constant over a long period of time) all soluble substances are liberated from the soil by weather processes and are being washed out. Since they are nowhere retained or accumulated, as would happen with an increasing mass of vegetation, they must appear in the ground, spring and creek waters. But since these waters are as chemically pure and poor as I have described, we must conclude that the soils are correspondingly poor and do not contain appreciable quantities of substances, including nutrients for plant growth, that could be mobilized by weathering. By implication this means that those soils are extremely infertile. Chemical analyses of soils (Camargo, 1948; Sombroek, 1966; Falesi, 1972) confirmed the conclusion drawn from the water chemistry. These results enable us to understand one fact or principle which is basic for the forest ecosystem of by far the greatest part of the Amazonian hylaea: the poorness in nutrients of its soils, or in other words their infertility. Recycling of nutrients