Technological advancements has made individuals and organizations more dependent on e-mails to communicate and share information. The increasing use of e-mails has led to an increased production of unsolicited commercial messages, known as spam. Spam classification systems able to self-adapt over time, with no human intervention, are rare. Adaptation is interesting as spams vary over time due to the use of different message-masking techniques. Moreover, classification models that handle large volumes of data are essential. Evolving intelligent systems are able to adapt their parameters and structure according to the data stream. This study applies the evolving methods TEDA (Typicality and Eccentricity based Data Analytics) and FBeM (Fuzzy Set-Based Evolving Modeling) for online unsupervised classification of spams. TEDA and FBeM are compared in terms of accuracy, model compactness, and processing time. For dimensionality reduction, a non-parametric Spearman-correlation-based feature selection method is employed. A dataset containing 25,745 samples, being 7,830 spams and 17,915 legitimate e-mails, is considered. 711 features extracted from an e-mail server describe each sample.