Root development is traditionally associated with the formation of Hertwig's epithelial root sheath (HERS), whose fragments give rise to the epithelial cell rests of Malassez (ERM). The HERS is formed by depletion of the core of stellate reticulum cells, the putative stem cells, in the cervical loop, leaving only a double layer of the basal epithelium with limited growth capacity. The continuously growing incisor of the rodent is subdivided into a crown analog half on the labial side, with a cervical loop containing a large core of stellate reticulum, and its progeny gives rise to enamel producing. The lingual side is known as the root analog and gives rise to ERM. We show that the lingual cervical loop contains a small core of stellate reticulum cells and suggest that it acts as a functional stem cell niche. Similarly we show that continuously growing roots represented by the sloth molar and K14-Eda transgenic incisor maintain a cervical loop with a small core of stellate reticulum cells around the entire circumference of the tooth and do not form a HERS, and still give rise to ERM. We propose that HERS is not a necessary structure to initiate root formation. Moreover, we conclude that crown vs. root formation, i.e. the production of enamel vs. cementum, and the differentiation of the epithelial cells into ameloblasts vs. ERM, can be regulated independently from the regulation of stem cell maintenance. This developmental flexibility may underlie the developmental and evolutionary diversity in tooth patterning.