When a commercial crop shows disease symptoms, it is important to make an accurate diagnosis to support control strategies. Diagnosing diseases in flowerbulbs requires considerable expertise. Only a few experts have the ability to do this, and each expert has his own specific domain. To retain expertise and to make it more generally accessible, an expert system, called EXSYS, has been developed. EXSYS contains disease descriptions and offers the ability to diagnose diseases in flowerbulbs. Digitalized photographic pictures of disease symptoms can be shown by the system to support an interacting session. This article describes the development of EXSYS, the structure of the knowledge base and the inference mechanism. A prototype has been built for diagnosing diseases in bulbous iris. Preliminary tests of this prototype at the Bulb Research Centre indicate that descriptions of observations can vary considerably between individuals. Adaptations have herefore been made to permit descriptions to differ from the exact descriptions in the knowledge base. Tests made it clear that a correct and complete knowledge base is necessary to create a successful system. The aim is to reduce the percentage of failures to less than 5%.