This study was conducted during 2003 and 2004 seasons at the Agricultural
Research Station of Gemmiza, to estimate optimum plot size and number of
replications ¡n fertilization and plant densities experiments for maize cultivars. The
experiment included 30 treatments which were the combinations of two maize hybrids
(Single cross 155 and Three way cross 327), Three plant densities ( 20000, 25000
and 30000 plant / fed) and five nitrogen levels (0, 40, 80, 120 and 160 Kg N /fed). A
split-split plot design with four replications was selected. The basic unit area was 10.5
m2. The grain yield data were recorded for each plot (Kg / plot). Data were subjected
to two procedures of statistical analysis , the first was exponential relationship (Smith;
1938) and the second was the maximum curvature (Lessman and Atkins; 1963).
The results can be summarized as follows:
1- Increasing plot size decreased variance per basic unit and coefficient of variability
(Ca V). However, the reduction was not in proportion with the increase ¡n plot size.
Index of soil heterogeneity ranged from 0.449 to 0.523 wìlh an average of 0.486
indicating moderate variation in the soil.
2- The exponential relationships between the coefficient of variability (C. V.) and plot
size (X) were:
C. V. = 7.907 X°647 for the first season.
C. V. = 13.804 X°7 for the second season.
3-The optimum plot size ranged from 0.815 to 1.096 basic units (8.557 to 11.512 m2)
for both seasons, respectively, by using Smith method. While it was 2.58 to 3.7
basic units (27.1 to 38.85 m2) by using maximum curvature method.
4- Increasing plot size and /or number of replications reduced the magnitude of the
difference detected at specified level of significance. The reduction of difference
with increasing plot size was less than that obtained by equivalent increase in the
number of replications.