Phosphorus benefits from grain-legume crops to subsequent maize grown on acid soils of southern Cameroon
MetadataShow full item record
Permanent link to this item: http://hdl.handle.net/10568/43946
We conducted field experiments over 2 years on two acid soils of southern Cameroon to test whether efficient uptake and use of phosphorus (P) from less available sources by grain legume genotypes could benefit subsequent rotational maize. We grew two crops each year. For the first crop we grew 4 genotypes of soybean and of cowpea, plus maize. For the second crop we grew maize. The first crops were fertilized with 0, 90 kg P ha?1 as phosphate rock (PR) or 30 kg P ha?1 as triple super phosphate (TSP). P application highly significantly increased shoot dry matter, P uptake, N2 fixation and grain yields of the grain legumes with TSP generally more effective than PR. Two of the soybean and two of the cowpea genotypes were more efficient at using P. Only the P-efficient soybean and cowpea genotypes increased subsequent maize yields. Yields of the subsequent maize grown in rotation were significantly correlated with shoot P uptake for which the quantity of P applied with the crop residues of the pre-crop appeared to be a major factor. We also grew the grain legumes in nutrient solutions and measured organic acid-anion exudation from roots, root-surface phosphatase-activity, and root morphological characteristics. Enhanced exudation of organic acid anions from roots of P-deprived plants might have contributed to the P acquisition efficiency under field conditions of the P-efficient cowpea genotypes and one of the P-efficient soybean genotypes. A higher activity of root-surface acid phosphatase might have been important for the other P-efficient soybean genotype. The results show, that the potential positive rotational effect of cowpea and soybean on the acid, highly P-sorbing soils of southern Cameroon depends on breeding and using P-efficient genotypes when sparingly soluble and suboptimal rates of soluble P fertilizers are used.