The phosphorus requirements of cereal crops with emphasis on the tropics

Abstract

The objectives of this study were: to determine the external and internal P requirements of four cereal crops (wheat, millet, sorghum and maize); to determine the relationship between P uptake by maize and the volume of soil fertilized at various P rates and to develop means for transferring soil management information from one site to another. In order to fulfill these objectives, the study was conducted in two parts. In the first part, the external and internal P requirements of four cereal crops, wheat, millet, sorghum and maize, were estimated from P field experiments identified in 5 countries. Soil samples and the field experimental data were secured. This was supplemented with experiments performed in Pakistan and in Hawaii. All the soil and plant analyses were done in Hawaii. Phosphorus sorption curves were developed for soils from each of the experimental sites. The procedure used to determine P sorption curves consisted of equilibrating 3 g samples of soil for 6 days at 25°C in 0.01 M CaCl2 containing graded amounts of Ca(H2P04)2. The samples were shaken for 30 minutes twice daily. Estimates of concentrations of P in solution established by ferti1ization in field were obtained from P sorption curves so developed. Crop yields were plotted as a function of the P concentration in solution. From these curves, the external P requirements were determined. For internal P requirements, leaf samples of an index tissue were taken from the experiments performed in Pakistan (Tandojam) and in Hawaii. In the second part, a series of three pot experiments were performed to investigate the influence of P rates, placement and mixing of a previously fertilized soil fraction with an unfertilized one on dry matter yield, P uptake and root distribution in maize. A P-deficient and high P-retentive soil (Tropeptic Eutrustox) was used. Maize was grown for 4 weeks in each experiment. The roots from fertilized and unfertilized soil fractions were collected separately and washed. Dry weights of shoots and roots were recorded. Phosphorus was determined in the shoots. The external P requirements of• four cereal crops varied within a narrow range from 0.018 to 0.025 ug/ml. For wheat grown at seven locations in Pakistan, India, Nebraska and Tanzania, the external P requirement determined from a composite yield response curve was 0.025 ug/ml P in solution (R2 = 0.7). At one location, two varieties of wheat were grown simultaneously and their external P requirements were 0.023 and 0.029 ug/ml. Millet data were obtained from two experiments, one in Hawaii with four varieties and the other in Pakistan with one variety. The external P requirement determined from a composite yield response curve was 0.018 ug/ml with R2 =0.78. For sorghum, experimental data were obtained from three experiments performed in Pakistan, Bangladesh and Hawaii. The external P requirement was 0.021 ug/ml as determined from a composite yield response curve (R2 =0.8). For maize grown on a Tropeptic Eutrustox in Hawaii, external P requirement was 0.018 ug/ml. At the other location (a Humoxic Tropohumult) in Hawaii, residual effect of P applied 2 years earlier was such that yield response to freshly applied P was small. It is likely that the decomposition of the bagasse (0.8% of the surface soil) added in the past, increased the avai1abi1ity of P. Thus the external P requirement of maize on this site was low (0.008 ug/ml). The internal P requirements of four cereals were: wheat, 0.25% P in recently mature leaf at ti11ering; maize, 0.3% P in ear-leaf at early si1king; millet, 0.29 to 0.42% P and sorghum, 0.37 to 0.41% P in flag leaf at head emergence. The results of greenhouse placement studies revealed that increasing rates of applied P resulted in increased shoot dry weights, root dry weights, P content and P uptake in shoots. With the lower rates of applied P, maize response was greater where P was mixed with small amount of soil. Shoot dry weights and P uptake were greater when P was mixed with 12.5% than when mixed with 100% of the soil volume. However, this trend reversed at high P rates so that localized placement of P was no more and often less beneficial than mixing P with entire soil volume. Subsequent study on the influence of mixing previously fertilized soi1 fraction with unfertilized soi1 fraction revealed that leaving the fertilized and unfertilized soil fractions unmixed, gave greater yields than mixing the soil fractions. The advantage of unmixed treatments over mixed treatments was greatest where a low P rate was previously applied to 12.5% soil volume. In case of mixed treatments, prior placement of low to moderate rates of P in a fraction of soil volume had some advantage over prior mixing of P with the entire soil volume. The results of this study suggest that P sorption curves can be used as a basis for transferring information about P fertilizer requirements from one site to an other irrespective of differences in P sorption capacity. The P sorption curves can be useful tool s for making predictions about P fertilizer requirements in areas where detailed experimentation has not been done. Where P fertilizer has been applied in the past the P sorption approach may underestimate the residual effectiveness of previously added P. When the amount of P fertilizer applied is less than that required for optimum yields, it may be advantageous to fertilize a small volume of soil.