This is an initial enquiry into the evaluation of hierarchical grouping procedures. The arbitrary criterion of quality proposed was goodness of fit of output distances to input distances. This was measured by two coefficients, Kendall's tau and a stress type of measure. Seven well-known hierarchical grouping techniques were investigated. The input data sets were artificially constructed to represent several different data types, including metric and ultrametric. A strong interaction was found between input data and grouping method.