The effects of high temperatures (46 to 5000) on the production of F pili by Escherichia coli were studied by electron microscopy. Attached F pili rapidly disappeared at 48 and 500C but not at 460C. Free pili were not denatured at these temperatures. The pili that disappeared from the cells at 500C did not appear as free pili in the culture supematant fluid, indicating that the pili had retracted to the cell surface or into the cell. The adsorption of either R17 phage or F pili antibody to the sides of pili prevented retraction. The disappearance of pili was accompanied by a loss in the ability to adsorb R17 phage but not M13 phage, suggesting that the tip of a pilus remains exposed after retraction.F pili rapidly disappear from the surface of donor strains of Escherichia coli K-12 under certain conditions. These conditions include infection with filamentous, donor-specific phages (9,11); incubation in the presence of sodium cyanide (13) and arsenate (17); and temperature shifts from 37 and 250C (14). Marvin and Hohn (11) were the first to suggest that F pili could retract. They observed that F pili disappeared during Ff phage infection and suggested that the pili retract into the cell with the phage on their tips. Other types of pili may also retract. The production of I pili by E. coli cells is stimulated by I pili antibody, and it was suggested that antibody might cause an increase in attached pill by preventing retraction (10). The shortening of Pseudomonas aeruginbsa pffi after adsorption of RNA phage to the sides suggests that they retract to the point of attachment of the phage (2). Novotny and Fives-Taylor (13) suggested that the disappearance of F pili in the presence of cyanide and at certain temperatures was due to retraction and presented a dynamic model for the growth of pili that supposes that F pili elongate and retract continuously and that the number of F pili per cell and the average length of pili reflect the state of equilibrium between these two processes.Studies of the outgrowth of F pili indicate that outgrowth requires the synthesis of ATP (16, 17) and the synthesis of RNA (8). The synthesis of DNA (16) or the synthesis of protein (3, 16) is not required. Although the synthesis of DNA is not needed for the reappearance of pili, there is evidence that DNA may be involved (7,8).Very little is known about the requirements for retraction. Retraction does not require the synthesis of ATP because pili retract in the presence of energy poisons (13, 17), and it does not require the synthesis of RNA and protein because rifampin, streptolydigin, and chloramphenicol have no immediate effect on retraction in the presence of cyanide (8). Retraction is inhibited by chloramphenicol after 15 min, suggesting that retraction requires a protein whose pool size is limited (8).Retraction and outgrowth are both sensitive to temperature. The sudden decrease in the number of attached F pili that occurs when cells are cooled from 37 to 250C (14) suggests that outgrowth, but not retraction, is inhibited at 250C. ...