Analysis of variance (ANOVA) implies a statistical technique on the variations to a group of causes. This .technique is closely related to *design of experiment *which may be defined as "the logical construction of the experiment in which the degree of uncertainty with which the inference is drawn may be well defined."

In a comparative experiment various objects of comparisons are termed as *treatments. *Also the whole experimental unit is divided into subgroups, preferably homogeneous are called *blocks. *

* *

Prof. R. A. Fisher has outlined the basic principles of the design of experiments which are

*(i) *Replication - means the repetition of the treatments under investigation.

*(ii) *Randomization - provides a logical basis for the validity of the statistical tests of significance.

*(iii) *Local control - which is a process of reducing the experimental error by dividing the heterogeneous area into homogeneous blocks and thus increases the efficiency of the design.