Cell lineage and cell states in the Drosophila embryo

Much of this essay is a cautionary tale about the inherent difficulty of building theories to explain pattern formation in embryos. In particular, I discuss the relationship between cell state and cell lineage. The idea that embryonic cells can occupy discrete genetic states is an old one but, because of the mass of new information from Drosophila developmental genetics, it has been given new life. I propose an operational definition of a cell state and argue that a true cell state should be propagated faithfully through cell lineage for at least some divisions. The genes involved should have a controlling role and the borders between populations of cells in two different states should be so sharp that cells fall clearly into one state or another.