BORDER EFFECTS IN A CONTROL PLOT IN A POT ASSIUM FERTILIZATION TRIAL ON EXHAUSTED SOILS IN THE IVORY-COAST

A factorial (4x2x2) fertilizer trial is being used for this study. It is more than 35 years old. The original design has been replicated on the second generation of plantings. The sandy soils were derived from tertiary sediments and the original K content was very low and homogen eous in the flat area. K application levels (4) have high significant effects on mean plot yields. A patch of 144 trees in 2 K0 level plots (control) and in surrounding plots receiving K fertilizer is being stu died. The individual data are collected for the following parameters: yield (weight, number of bunches), mineral nutrition (major elements) and growth (stem height, area and number of green leaves). The soils a re analyzed. Mapping of the different parameters shows different gradi ents due to the ''poaching'' effect. Thus, the conclusions draw agrono mists' attention to the plot result studies and the plot size determin ation.