Abstract This paper provides an overview of all the meaningful sub-sign form units (form-meaning units; FMUs) in lexical signs in Sign Language of the Netherlands (NGT). We investigated the potential meaning of all form features that were previously established in analyses of NGT form by analyzing their distribution in lexical signs. The data set consisted of 500 NGT signs in the lexical database Global Signbank, and a set of 163 elicited newly-formed lexical signs. All features in these data sets appear to bear meaning (at least once). No completely arbitrary features were found, and some features appeared to be always associated to a specific meaning. This toolkit and the set of FMUs in NGT provides a possible basis for cross-linguistic study and for a more fine-grained approach in various research disciplines, for instance psycholinguistics and acquisition, and it may thus advance the theoretical and applied study of sign languages.