The enormous losses suffered by the European elms during recent Dutch elm disease outbreaks led to concern over the conservation of elm genetic resources, and the subsequent establishment of a series of ex situ collections. However, as ex situ collections are inevitably finite in size, some consideration needs to be given to selecting which samples to include in them. To contribute towards this process for European ex situ elm collections we have undertaken genetic studies on a Europe-wide sample of 535 individuals. A major aim has been to use genetic markers to clarify the identification of samples to ensure that the ex situ collections contain a representative spread of taxonomic diversity. This is important given the paucity of mature elms in the landscape due to Dutch elm disease. The lack of mature material (critical for identification) compounds identification problems in what was already a taxonomically difficult group. Our data (derived from random amplified polymorphic DNA and inter-simple sequence repeats) have provided a useful supplement to morphology in undertaking such sample identifications. The molecular data served to highlight mis-identified samples and led to extensive revisions of sample identities within individual countries. Our results were less useful in detecting regional intra-specific genetic structure, and do not provide sufficient information for prioritising within-species sample selections.