The definition and recognition of silver minerals is very difficult. In general such minerals occur at low concentration and can be hard to identify and measure using conventional optical mineralogy. The present paper describes the use of the QEM*SEM system to tackle this problem and provide information of use to the metallurgist. Three aspects of QEM*SEM are used to special advantage in this application. Firstly the automated mode of operation is used to make a rapid search for the sparse phases. Secondly the X-ray capability is used to identify the mineral phases. Finally the image analysis programs quantify the distributions of the minerals which are found. The paper deals with the identification stage of the process and with the use of the data which can be derived. Silver minerals which covered a range of ore types and composition were collected and studied in detail. On the basis of these measurements and information received from a number of mining companies regarding the important minerals, a classification table was prepared. This table listed about 40 different silver mineral types which could be identified by QEM*SEM and which were of interest in silver recovery processes. Where the silver occurs in solid solution in concentrations greater than about one per cent, the silver content in the species was subdivided into ranges based on statistical intervals. The QEM*SEM system was programmed to recognise these as separate species. Examples are presented of the use of the system in defining the type and morphology of the silver mineralisation to assist in the assessment of batch flotation test Samples.