SHORT-TERM AND LONG-TERM EFFECTS OF CROP RESIDUES AND OF PHOSPHORUS FERTILIZATION ON PEARL-MILLET YIELD ON AN ACID SANDY SOIL IN NIGER, WEST-AFRICA

Field experiments were conducted during the rainy seasons of 1990 and 1991 on an acid sandy soil (Luvic Arenosol) in Niger, to assess long-t erm (since 1986) and short-term (since 1990) effects of millet straw ( crop residues) at different amounts (2 t and 6 t ha-1) and modes of ap plication (incorporation, mulching and burning), and of phosphorus (P) fertilization on dry matter yield of pearl millet [Pennisetum glaucum (L.) R. Br.]. Especially long-term, but also short-term application o f crop residues increased dry matter yield of pearl millet by more tha n 60%, whereas their omission decreased yield immediately. Compared to mulching or burning, higher yields were obtained with incorporation o f crop residues. When crop residues were applied as mulch, dry matter yield was increased to the same extent by 2 t and 6 t ha-1. Similar dr y matter yields were achieved with crop residues (-P) to those with P fertilizer only. An additional yield increase could be obtained by sim ultaneous application of crop residues and P. Without crop residues, p otassium (K) concentrations in the shoot dry matter indicated K defici ency. With crop residues considerable amounts of K (15 kg t-1) were pr ovided and raised the K concentrations in the plants to the sufficienc y range. Although with crop residues some P was provided ( 1.5 kg t-1) , the beneficial effects were primarily attributed to increased P acqu isition by the millet plants, which was reflected in enhanced root gro wth in the topsoil (0-10 cm).