PENIMAT is a computer assisted probabilistic identification scheme for terverticillate species of the genus "Penicillium". The identification is made up from up to 57 characters derived from morphological, physiological, biochemical and scanning electron microscopy methods. This identification package consists of a data file with a frequency matrix giving the frequency of each of the 57 attributes in 37 species or part-species groups. The attributes of an unknown isolate are then compared against each of these species groups and a quantitative measure of the likelihood of the unknown isolate belonging to each group is calculated. In addition, a second measure relating to the difference between the unknown isolate and the most typical member of each group is also calculated. When these quantitative measures exceed inbuilt threshold values, an identification is given to the unknown isolate. This PENIMAT system allows direct comparison between the attributes of the unknown isolate and the most likely identification, lists the attributes that differ between them and provides a written report of the identification.