Introduction: An important goal for understanding how animals have evolved is to reconstruct the ancestral features and evolution of the nervous system. Many inferences about nervous system evolution are weak because of sparse taxonomic sampling and deep phylogenetic distances among species compared. Increasing sampling within clades can strengthen inferences by revealing which features are conserved and which are variable within them. Among the Annelida, the segmented worms, the Clitellata are typically considered as having a largely conserved neural architecture, though this view is based on limited sampling. Results: To gain better understanding of nervous system evolution within Clitellata, we used immunohistochemistry and confocal laser scanning microscopy to describe the nervous system architecture of 12 species of the basally branching family Naididae. Although we found considerable similarity in the nervous system architecture of naidids and that of other clitellate groups, our study identified a number of features that are variable within this family, including some that are variable even among relatively closely related species. Variable features include the position of the brain, the number of ciliary sense organs, the presence of septate ventral nerve cord ganglia, the distribution of serotonergic cells in the brain and ventral ganglia, and the number of peripheral segmental nerves. Conclusions: Our analysis of patterns of serotonin immunoreactive perikarya in the central nervous system indicates that segmental units are not structurally homogeneous, and preliminary homology assessments suggest that whole sets of serotonin immunoreactive cells have been gained and lost across the Clitellata. We also found that the relative position of neuroectodermal and mesodermal segmental components is surprisingly evolutionarily labile; in turn, this revealed that scoring segmental nerves by their position relative to segmental ganglia rather than to segmental septa clarifies their homologies across Annelida. We conclude that fine taxonomic sampling in comparative studies aimed at elucidating the evolution of morphological diversity is fundamental for proper assessment of trait variability.