The role of specific morphogens is well established in the determination of body plans in development. A variety of morphogens have been identified; others are suspected. Pathways have been delineated. In complex tissues, the ability to maintain fidelity of microarchitectural structure is crucial. Microarchitecture is a consequence of relationships among cells, not a function of single cells. Epithelial layers, in particular, are able to maintain their microarchitecture with remarkable accuracy over many decades despite recurrent damage, regular cell turnover, and complexity of structure. Nonetheless, metaplasia and transdifferentiation (change in tissue structure without cell dysplasia) do occur, suggesting that there is the possibility of loss of control or change of control of the microarchitecture. A strong inference to be derived from the above is that there are control systems and molecules and that these are derived from cells that are outside, but plausibly adjacent to, the respective epithelia. It is postulated is that there are morphogen-like controller molecules with morphogen-like functions in adult epithelial tissues. These are responsible for the maintenance of normal tissue microarchitecture. Because the function of these putative molecules is maintenance of tissue structure, I have chosen to call them morphostats by analogy with morphogens. It seems plausible that morphostats and morphogens may constitute overlapping families of molecules. Evidence for the existence of morphostats can be derived from a variety of in vivo and in vitro data and from studies of normal tissue, precancer, and cancer, including: (a) the existence but rarity of metaplasia and transdifferentiation; (b) the fact that metaplasias are multicentric and are only one step from normal but do not show any consistent epithelial mutation; (c) the genesis of animal cancers by simple transplantation of tissues into the wrong environment and the evidence that epithelial mutation is not a feature of such transplantation carcinogenesis; (d) the fact that carcinogenesis occurs frequently at the junctions of different epithelial types, e.g., squamocolumnar junctions in gastrointestinal and genital tracts; (e) the fact that cancer-associated fibroblasts can stimulate proliferation in transformed cells but not influence normal cells; and (f) the failure to grow most epithelial organs in a fully differentiated structural pattern in monolayer culture. It is suggested that morphostats may function like morphogens inasmuch as they may act via a diffusion gradient from source mesenchymal cells and provide architectural instruction for complex adult epithelia. Morphostats may influence architecture via control of cell adhesion, apoptosis, and proliferation. Some specific predictions follow from this hypothesis, most notably, a new two-hit model of cancer: one mutation in an epithelial cell resulting in disruption of cell function and structure (e.g., dysplasia); and the other in a mesenchymal or other supporting cell resulting in disruption of tissue microarchitecture. The corollary of this is that there will be mesenchymal mutations producing microarchitectural abnormalities without epithelial dysplasia and vice versa. Disruption of the functions of morphostats may result in a variety of abnormalities. Such disruption may be a key event in carcinogenesis.