A diverse set of direct experimental observations and theoretical considerations supports the existence of fluidity in cell membranes (1) . Several membrane-dependent functions of Escherichia coli have been shown to exhibit biphasic Arrhenius plots with transition temperatures that are determined by the degree and configuration of the unsaturation of fatty acids incorporated into the membrane phospholipids (2, 3) . Biphasic plots and even discontinuities in the plots have been demonstrated for several mitochondrial enzymes in their membrane-bound states (4, 5) . It has been proposed that the transitions in activation energy for these reactions are related to a phase change in the organization of the membrane phospholipids (2-5) . In mammalian cells the mixing of surface antigens in newly formed heterokaryons (6) and the capping of surface antigens in the presence of specific antibodies (7) have been interpreted as evidence for lateral movement of proteins in a more or less fluid membrane . Both these phenomena have a temperature dependence that is consistent with a transition in membrane structure occurring in the vicinity of 15°C, but the available data do not permit distinction between phase transitions and high activation energies for lateral diffusion of membrane proteins (7) . Secretion by exocytosis would be expected to be dependent on the state of the cell membrane and thus substantially affected by any temperaturedependent phase change in the membrane. This reasoning led to study of the effect of temperature on the secretion of histamine by mast cells .