statistical treatment of the data from that design,
requires statistical and technological expertise.
Surprisingly often a form of analysis for a completely
crossed design has been used despite the description of
the experiment clearly demonstrating that this should
not be the case.