We propose a novel solution to the email classification problem: the integration of temporal information into traditional content-based classification approaches. A lot of research has been recently carrying out on email classification. Generally, those approaches concentrate on content interpretation and work only with content-based features. At the same time, email data have a temporal character. Every email message has a timestamp, the time it was received by a server. This temporal information is always present but usually ignored by researchers. We would like to explore the potential relevance of temporal information to common tasks in the email domain. We discover temporal relations in an email sequence in the form of temporal sequential patterns and embed the discovered information into content-based learning methods. The performance of the new heterogeneous classification system has been tested on several email and newsgroup datasets. The system has been able to reduce the classification error by up to 22%.