Designs for sets of experimental units with many blocking factors are studi
ed. It is shown that if the set of blocking factors satisfies a certain sim
ple condition then the information matrix for the design has a simple form.
In consequence, a design is optimal if it is optimal with respect to one p
articular blocking factor and regular with respect to all the rest, in a se
nse which is made precise in the paper. This encompasses several previous r
esults for optimal designs with more than one blocking factor, and applicat
ions to many other situations are given.