Diagrammatic representation of how a multivariate approach to morphological study of shell form for cowries was applied in this study, using Umbilia armeniaca as an example (see Suppl. material 1 for an outline of code used) A data were sourced by examining unstudied specimens (primary data) or sourced from prior studies (secondary data) B data for morphometrics deemed representative of shell form were C transformed to Z-scores and atypical specimens either validated (primary data) or censored (secondary data) before D computing a resemblance matrix based on Euclidean distance between specimens E non-metric multidimensional scaling (nMDS) was then used for dimensionality reduction to F permit visualisation in a new space of two dimensions and G statistical testing was employed to validate visual observations by estimating the probability that a priori assigned groups (taxa or populations) shared the same centroid and dispersion within multidimensional space.