In an analysis of multitrait-multimethod matrices the criteria for discriminant validity are shown to include a "structure" criterion as an invariance of traits structure to methods. The criterion is meant to fit data to an additive model with traits and methods but not interaction terms. The importance of the structure criterion and the relative unimportance of the two elevation criteria are shown. A fourth criterion which is also a "structure" hypothesis is discusses as a "heterotrait-heteromethod" block. Applying the facet approach to the analysis of multitrait-multimethod matrices is a step forward in validity investigations since it ties together Campbell and Fiske's criterion which is "theory free" with Cronbach's ideas about construct validity. The pictorial nature of space diagrams helps researchers see many aspects of data usually unnoticed in other approaches. (Author/CM)