Simple leaves are usually conceptualized as terminal organs that cannot develop new roots or shoots. However plants with succulent or semi-succulent leaves in many unrelated families defy this rule. They do so by differentiating leaf parenchyma — which are probably often phloem parenchyma or companion cells, the two of which are often indistinguishable — into new root and shoot apical meristems. Succulent and semi-succulent leaves can survive longer than non-succulent leaves once detached from shoots of their parent plant, possibly enabling them to differentiate and establish a root system before desiccation and/or starvation. Curiously, almost no succulent plants have compound leaves, even though compound leaves are often considered to have properties of both stems and leaves and hence theoretically should have the capability of developing new structures, such as new root and shoot meristems. I discuss some testable predictions arising from these notions and ask whether differences between ability to regrow roots and shoots from detached leaves of monocots versus eudicots is better explained by reticulate versus parallel venation or phylogenetic constraints.