Pasture development has become the largest anthropogenic disturbance of forest land in the Amazon basin (Skole et al. 1994, Serrão and Toledo 1990). The area of forests converted to cattle pasture in Amazonia is currently estimated at approximately 20 million hectares. In the Brazilian Amazon basin, most of the conversion of forest land to pasture began during the early 1960s to the late 1980s, as a consequence of the opening of Amazon highways and government policies aimed at regional development (Hecht 1982, Nepstad et al. 1991, Serrão et al. 1979). Pasture productivity and longevity in the Amazon basin seem to be closely related to soil fertility and nutrient cycling (e.g., Dias Filho and Serrão 1987, Serrão et al. 1979). Thus, understanding the major biogeochemical cycles that influence soil fertility under pasture is vital for predicting the consequences of continued conversion of tropical forests to cattle pastures. This understanding is also important for devising management technologies that enhance the sustainability of these areas and thus slow further deforestation. Although during the first three to five years after establishment, the productivity of pastures is often good, after that period a rapid decline in productivity of the planted grasses associated with an increased presence of herbaceous and woody invaders is generally observed (reviewed by Serrão and Toledo 1990). If left uncontrolled, these invader species slowly become dominant and lead to “pasture degradation,” a condition characterized by a complete dominance of the weedy community. If left to secondary succession, forest vegetation usually becomes reestablished on these degraded pasture lands in the Amazon, although the species composition is usually different than that of the primary forest (Nepstad 1989). The nutrient status of the degraded pasture soils is among the factors that affect the rate of regrowth of the secondary forests. One of the first attempts to study soil nutrient dynamics under cultivated pastures in the Amazon basin was conducted in the early 1970s by Falesi (1976). The results of that chronosequence study in different soil types suggested that soil nutrient cycling in pastures differed from that of the traditional slash-and-burn agriculture.