Fodder beet is an ideal fodder crop for dairy cattle. As border and end-
border effect (alley width and end plants of the rows) is one of component of the
experimental error of fodder beet trials. Removing the rows, which are adjacent to the
alley and end plants of the rows, can eliminate this effect. The purpose of this study
was to develop accurate factors to adjust rows yields to decrease sampling errors of
fodder beet trials to control border and end-border effects in test plots. Two field
experiments were conducted at Sids Agricultural Research Station Beni Suef
Governorate, during the two growing seasons of 1999/2000 and 2000/2001 seasons.
Four alley width were 0.7, 1.40, 2.10 and 2.80 m used as a treatments to clear the
border effect and it arranged randomly in the middle of each plots in randomized
complete block design with four replications. To study the border effect (alley width)
on the yield of six rows from each side of each alley was calculated separately. The
statistical analysis included all 12, 10, 8, 6, 4 and 2 rows after discarding 2, 4, 6, 8
and 10 rows adjacent the each side of the alleys, respectively. For study the end-
border effect (end plants in rows) the root weight of each single plant in each row (12
plant/row) was calculated for all experimental plots. The statistical analysis included
all 12, 10, 8, 6, 4 and 2 plants in each row after discarding 2, 4, 6, 8 and 10 plant from
each end of the all rows in the plot, respectively. The results for this study could be summarized as follows: I-Effect of border distances (alley width) in fodder beet experiments: The effect of removing the first and the second row from each side of each
border distances in average of fresh fodder beet yield/plant, sampling error,
experimental error and coefficient of variability for the efficient of the experiment and
the accuracy of the anaiysis could be summarized as follows: 1- The average of fresh fodder beet yield/plant was reduced from 2.705, 3.337,
3.651 and 3.841 to 2.171, 2.268, 2.278 and 2.313 in the first season and it
reduced from 2.649,3.087,3.433 and 3.768 to 2.209, 2.296, 2.278 and 2.250 in
the second season. 2- The sampling error was reduced from 5.641 to 0.227 in the first season and from
6.178 to 0.278 in the second season. 3-The experimental error was reduced from 5.884 to 0.235 in the first season and it               reduced from 4.709 to 0.169 in the second season.      . 4-The coefficient of variability was reduced from 20.31 to 7.46% in the first season
and it reduced from 22.18 to 8.26% in the second season. This result indicated that border effects extended to the first and the
second row for all plots. Results also indicated that the efficient of the experiment and
the accuracy of the analysis was increased and removing the first and the second row