A simple derivation of expected mean squares is given for the randomiz
ed (complete) block design, showing that ''experimental error,'' the e
rror term for testing treatments, is comprised of three sources of var
iability: block by treatment interaction, within block plot-to-plot va
riability, acid within experimental plot sampling variation. The appro
ach could readily be extended to incorporate measurement error as a fo
urth component of experimental error.