SUMMARY: Two strains of bacteria which produced flagella when grown at 36° but not at 44° were examined; one was a strain of Salmonella typhimurium and the other of Proteus vulgaris. These organisms were grown on membranes for electron microscopy, being incubated at 36° so that the parent bacteria of each microcolony possessed a normal quota of fiagella, and then transferred to 44° so that no more flagella were produced. In the microcolonies, after several divisions, it appeared that all these flagella were retained by the original parents, in accordance with the theory that bacteria of this type divide by budding from a growing-point at one pole.