In many experimental situations the pertinent inferences are made on the basis of orthogonal contrasts among the treatment means (as in 2 n factorial experiments). In this setting a particularly ...