Abstract:

Declining soil fertility is a fundamental impediment to agricultural growth and a major reason for
the slow growth in food production in sub-Saharan Africa. Among the primary plant
macronutrients, phosphorus is the most important nutrient element, limiting agricultural
production. In order to manage soil phosphorus objectively, the existing stocks in the soil should
be known. It is on this basis that informed effective recommendations for intervention can be
made. However, among the challenges in achieving reliable information lies the use of the most
efficient analytical methods that simulate closely with plant nutrient uptake and growth. The
most widely used soil P extractants are Bray I, Bray II, Olsen, Mehlich I, II and III. These
methods were largely developed targeting temperate soils and eventually extended for use under
tropical soils without the necessary evaluation for relative effectiveness and efficiency. This
study was. thus, geared to evaluating the common methods of testing for soil available P in
laboratories in Uganda so as to identify the most appropriate method. Treatments administered
included six P rates namely, 0 (Control), 20, 40, 60, 80 and 100 kg P ha'1. Three soil phosphorus
extraction methods used included Bray I, Bray II and Mehlich III. The experiment was set in a
greenhouse at Abi Zonal Agricultural Research and Development Institute, Arua, Uganda. Maize
variety Longe V was used as the test crop. Data collected were analysed with the GenStat
software, and correlations were run between available P fractions extracted using the different
methods against dry matter and plant growth parameters. Mehlich III presented the highest
correlation coefficient with stem girth (r = 0.63), number of leaves (r = 0.733) and plant P (r
0.254). There was a strong positive correlation between Bray 1 and Mehlich III methods (r =
0.975), thus either could be used to quantify extractable P in Ferralsols with mica, kaolinite and
quartz mineralogy of 1.57% organic carbon and pH 5.6.