CROPPING AND FALLOWING SEQUENCES OF SMALL FARMS IN THE TERRA FIRME LANDSCAPE OF THE BRAZILIAN AMAZON - A CASE-STUDY FROM SANTAREM, PARA

This paper analyzes field survey results and develops a conceptual mod el of the factors that influence cropping and fallowing practices on s mall farms in the terra firme landscape near Santarem, Brazil. A multi -fallow cultivation system that used rice, corn and bitter manioc in v arious relay-intercropping combinations was the most common cultivatio n practice observed. Five different types of fallow vegetation were id entified and used by the farmers: (1) mature forest vegetation greater than 20 years old; (2) secondary forest vegetation 8 to 1? years old; (3) young secondary forest vegetation 3 to 6 years old; (4) brushy ve getation 2 to 4 years old; and (5) weed vegetation less than 2 years o ld. Distinct relay-intercropping sequences were associated with each o f these fallows. We suggest that the selection of fallow length and cr opping sequence is subject to the following general constraints: (1) t he productivity of the landscape as determined by soil, water and clim ate; (2) ecological requirements and risks associated with particular crops; (3) land availability and the costs of site preparation, and cu ltural treatments; (4) the availability of hired labor; (5) the age st ructure of the families, their subsistence requirements and preference s for particular crops, leisure and non-farm-related production activi ties: and (6) local economic conditions including land values, access to credit and non-farm-related employment, and the conditions of commo dity markets. To maximize agricultural production and general househol d utility given these constraints, the farmers have several options, i ncluding: (1) varying the length of fallows, (2) varying the types and sequences of crops that are planted following a given fallow; (3) mod ifying the clearing and cultivation practices; (4) improving subsequen t yields by managing regeneration within a fallow; (5) developing dive rsified land use systems that contain combinations of pasture, perenni als, semi-permanent annuals, areas of extractive reserves and true shi fting cultivation; and (6) increasing production through the use of ex ternal inputs like fertilizer, irrigation and farm machinery. Most far mers in the study area have chosen to modify cropping sequences and va ry the lengths of natural fallows rather than using expensive external inputs. Our data suggest that a major factor in selecting a fallow le ngth is the cost of land clearance and preparation. Moreover, since cl earing costs are dramatically reduced for young secondary vegetation, the reduction in site preparation costs over several short rotations c ompensates for the lost production caused by using short fallows inste ad of long fallows.