EFFECT OF LIMING, N AND P FERTILISATION OF A LIXISOL ON THE GROWTH OF SELECTED SOYBEAN CULTIVARS UNDER SUB-HUMID TROPICAL CONDITIONS IN ZIMBABWE - Soybean and Wheat Crops: Growth, Fertilization, and Yield - page 88

banded into planting rows at planting as ammonium nitrate (AN) (34.5% N) and single super
phosphate (8% P and 12% S), respectively.
In both experiments lime was broadcasted on the soil surface and incorporated into the 0-
15cm soil horizon. Plots were 5m x 5m in size and planting rows were marked out using hand
hoes. Soybean seed was inoculated with Bradyrhizobium strain MAR 1491 (USDA 110)
(Kasasa et al., 1998) in both seasons. Seed was banded in rows 45 cm apart and plants were
then thinned to 7 cm intra-row at two weeks after germination to give a target plant
population of 320 000 plants ha -1 . Weed control was done by hand-hoeing throughout the
cropping season.
2.3. Assessment of Nodule Number and Effectiveness
Nodulation was assessed at eight weeks after planting by carefully digging up root
systems of ten randomly selected plants in each plot. A 3x3m net plot at the centre of each
plot was reserved for yield determination at harvesting. The nodules were removed from the
roots, dried and weighed. To determine nodule effectiveness, the internal colour of five fresh
nodules per plant was recorded. A red colour depicted an effective nodule, pink colour a
weakly effective nodule, and green or white colour depicted an ineffective nodule (Zengeni,
2004).
2.4. Grain and Stover Yield Determination
Grain and stover yield were determined at physiological maturity (16 weeks after
planting) using plants in the 3mx3m net plot. Total aboveground dry matter yield was
determined by weighing, and after threshing and winnowing grain yield was determined by
weighing and adjusted to 11% moisture content. Stover yield was obtained by subtracting
grain yield from total aboveground dry matter yield. Leaf litter was not included in the stover
yield as the leaves had fallen to the ground by the time of harvesting. In the second
experiment grain and stover sub-samples were taken for N and P analysis (Anderson and
Ingram, 1993).
2.5. Data Analysis
Analysis of variance (ANOVA) using Genstat (2002) was used to identify treatment
effects. Variables on which statistical analyses were performed are stover and grain yield,
nodule numbers and effectiveness. Least significant differences (P<0.05) were used to
separate treatment means. The effects of lime, cultivar, N and P were quantified as well as
interaction effects.