The term mesosome introduced by Fitz-James has generally been used to describe invaginations of the microbial cell membrane which are not enriched with photosynthetic-associated pigments. If mesosome structures were related to normal physiology, one might expect these structures to be observed at site-specific locations within the cell, and to appear at discrete times in the cell cycle; to be altered in number or size by manipulation of parameters related to the cell cycle; and to show differences in chemical composition when compared with the cytoplasmic membrane. As a working model it is proposed that both peripheral and central mesosomes are produced by the invagination of a portion of the cytoplasmic membrane at sites which may have a lowered affinity for the cell wall. These membrane sites would be pulled into the cell after a perturbing event to form mesosomes as a result of a differential contraction of the cell-wall or cytoplasmic volume.